Sexual Behavior

views updated

Sexual Behavior



II. SOCIAL ASPECTSSamuel Z. Klausner









Animal sexual behavior is but one aspect of the total pattern of reproductive behavior. Sexual behavior includes all responses directly associated with genital stimulation and copulation, whether homosexual or heterosexual. Among warm-blooded vertebrates, heterosexual copulation culminates in the transfer of sperm from male to female, with subsequent fertilization of ova and reproduction of the species.

Despite its significance for species survival, sexual behavior has received relatively little experimental attention. With few exceptions, sexual behavior in man has not been studied experimentally and is likely to remain generally inaccessible for study by experimentalists for some time to come (Ford & Beach 1951; Lloyd 1964). Careful experimental analysis of animal sexual behavior, especially that of primates, should provide a valuable background of information for understanding the data that exist concerning sexual behavior in man. Many of the forms of sexual behavior described by Kinsey and his associates (Kinsey et al. 1948; 1953) as part of the human sexual behavior repertoire have been observed in a number of other mammalian species. Beach (1949) has suggested that a study not only of the similarities but of the differences in sexual behavior between man and other mammalian species should help in our understanding of how social forces can modify a basic biological drive.

In experimental research on sexual behavior, as in research on any other aspect of behavior, the experimentalist seeks to discover and analyze those variables of which behavior is a function. The initial step in the process is the identification of components of sexual behavior and a description of these in relation to the total pattern. It then becomes possible to measure quantitatively the latency, frequency, and duration of each component and thus to assess the effects of the independent variables on specific components of the total sexual behavior pattern.

Since there have been many reviews of sexual behavior patterns of a number of insects, fishes, amphibians, reptiles, and birds, the present account is restricted to a discussion of sexual behavior in placental mammals. Detailed descriptions are available of the mating behavior of several laboratory mammals, including the rat, guinea pig, hamster, rabbit, cat, dog, monkey, and chimpanzee, and of farm animals such as the goat, sheep, cow, pig, and horse. Limited descriptions are also available of sexual behavior patterns in normally wild species not generally found in laboratories, such as the chinchilla, porcupine, skunk, mink, fox, and elephant. The accuracy and detail of the descriptions of the behavior in the last-mentioned species vary widely, however, and further study is needed.

Components of sexual behavior


The most readily identified aspect of animal sexual behavior is the mounting response. It is an essential component of the male mating pattern but frequently occurs in females, especially during estrus. In a successful mating sequence the male mounts the female from the rear. The forelegs are placed on the back or around the sides of the female with the hind legs on the ground. The male may then execute a series of rapid pelvic thrusts. The posture and motion permit the male to insert the erect penis into the female’s vagina. Not all mounting, however, is oriented to the rear of the female. The male may mount the side or head of-the partner and may or may not execute pelvic thrusts. Mounting by females also shows similar variation. Females of a number of species display mounting behavior only during estrus; but in some species, such as the rat and dog, mounting by females is not restricted to any particular part of the estrous cycle. The mount executed by a female is frequently indistinguishable from that of the male. Thus, the typography of the behavior does not identify the sex of the animal. Mounting occurs between members of the same or opposite sex and across species in some instances.


Intromission is obviously restricted to the male. To achieve intromission, not only must the male be properly oriented, but the penis must be erect. The detection of intromission is difficult in most species and often requires special conditions of observation. In some species, such as the guinea pig, the rate of pelvic thrusting may be slower during intromission. In the rat, when intromission is not accompanied by ejaculation, the intromission response is usually followed by a vigorous backward lunge and a period during which the male may manipulate the penis with paws and mouth before mounting again. This gross bodily reaction is most frequently used as an index of intromission rather than visualization of penis insertion.


Ejaculation, as in the case of intromission, is not identified directly but is commonly inferred by changes in gross body movement. In the guinea pig, for example, at the moment of ejaculation the flanks of the male are drawn in as in a spasm. The male then dismounts, or the female pulls away, and both male and female clean the genitalia. Ejaculation by the male rat is also accompanied by a drawing in of the flanks. Following ejaculation the male dismounts much more slowly than he does following intromission, and there ensues a period of relative inactivity. Judging whether or not ejaculation has occurred from observation of gross behavior is particularly difficult in some species, and the presence of sperm or an ejaculatory plug in the vagina is frequently used as confirmatory evidence.

Lower components

Not only is mounting the most conspicuous aspect of sexual behavior but together with intromission and ejaculation shows for most species the least variation among the numerous components of the sexual behavior pattern. Wide variation exists from species to species in the so-called lower components of the mating pattern. In general, these lower components appear to produce excitement and readiness to mate in both sexes. The guinea pig displays a mating pattern typical of many species. The following sequence is characteristic: On encountering an estrous female, the male, walking with a characteristic gait, circles the female; he rubs against her flanks, nibbles at the ear, prods the female’s mouth area with his nose, licks the genitalia, then approaching from the rear, rubs his nose and lower jaw over the female’s back, a response that frequently elicits the lordosis response in the receptive female.

These lower components may occur in any order and do not represent a stereotyped sequence of reflexes triggered by the presence of an estrous female. In some instances, especially in experienced individuals, the male may mount the female without any preliminary display and may achieve intromission and ejaculation without manifesting any of the lower components.


The one outstanding component of female sexual behavior common to all mammalian species except man is the lordosis response. In the monkey the receptive posture is referred to as a “present.” The receptive posture or lordosis consists of a particular stance, with the four legs fixed and a straightening and arching (in some cases) of the back with elevation of the pudendum. In some species the head is thrown backward and the tail is deviated or displaced so as to expose the genitalia. A number of minor species differences exist, but in all species the pattern facilitates both mounting and intromission by the male.

The complete pattern of female sexual behavior may include a number of other components in addition to the lordosis response. These additional components vary widely among species. They may precede copulation and may be classed as courtship behavior or “foreplay” or precede the onset of full estrous behavior as in the guinea pig and be referred to as proestrous behavior. In some species, such as the cat, estrus may be followed by distinct behavioral responses, termed afterreaction.

Factors influencing sexual behavior

Accurate description of sexual behavior in natural surroundings and under carefully controlled laboratory conditions, although an essential first step, constitutes but a starting point for the study of the major problems associated with sexual behavior. In experimental investigation, sexual behavior has been found to vary with the species, strain, sex, diet, temperature, illumination, season, age, method of rearing (including age of weaning), nature of housing (including social and isolated conditions), stage in the reproductive cycle, familiarity with the test area (territoriality) and test partner, duration and frequency of tests, dominance relationship, and the general health of the animal. With the exception of masturbation, sexual behavior is a social phenomenon; hence the behavior of each individual of a pair or group constitutes an added stimulus variable not readily controlled in the assessment of sexual behavior. It is frequently convenient to classify the determinants of sexual behavior into a few broad categories. Such categories are usually neither all-inclusive nor mutually exclusive but suggest, in a general way, the class of variable emphasized by an investigator or represented in a given research effort. For purposes of presentation we may categorize the variables as genetic, hormonal, sensory and neural, and experiential.

Genetic variables

Chromosomal sex is determined at the time of fertilization and depends upon the particular chromosomes contributed by the parents. Many problems of sex determination at the gene level of analysis remain to be solved, but these need not concern us here. Classification of the individual as male or female is usually based on the appearance of the external genitalia. Unfortunately sex classification based on this criterion does not always correspond to chromosomal or gonadal sex, nor does it necessarily correspond to the pattern of sexual behavior displayed by the animal. In addition to these possible inconsistencies, chromosomal sex itself may not be normal, as for example in the case of an individual whose chromosomal pattern is XO and who, accordingly, lacks one of the sex chromosomes. Nevertheless, species differences, strain differences, and sex differences may be changed by varying genetic background. Such knowledge contributes little to our understanding of how the genes bring about their effect; but regardless of the mechanisms involved, it is assumed that all organismic variables have a genetic basis (Beach 1947a).

Breeding experiments

The genetic material contributed by the parents has been shown to determine the particular pattern of sexual behavior that will be displayed by the individual as an adult. Not only is it possible to breed for high or low levels of sexual behavior, but different modes of inheritance of the various components of sexual behavior have been demonstrated. For example, in a study of two inbred strains of guinea pigs, in which the male of the two strains differed in ejaculation latency and in mounting frequency, Jakway (1959) has shown that ejaculation latency of one strain was dominant and mounting frequency of the second strain was dominant when these were studied in F1 and F2 hybrid generations and in backcrosses. Differences have also been demonstrated in modes of inheritance of components of female sexual behavior in the guinea pig (Goy & Jakway 1959). It is obvious that sexual behavior need not be inherited as a unitary trait. Although most work on the inheritance of patterns of sexual behavior has been carried out on insects, an increasing amount of work is being done on mammals (Dilger 1962; Goy & Jakway 1962).

Hormone variables

Sexual behavior among females of lower mammalian species is characterized by its cyclicity. The display of sexual behavior is referred to as estrus or heat. The period coincides with the presence of mature ovarian follicles and, generally, with changes in the vaginal epithelium. The term estrus is sometimes used to refer to the physiological state of the ovary or condition of the cells in the vagina. The multiple meaning attached to the term is testimony to the close relationship between the physiological state and the behavioral state. Ovulation usually, although not always, accompanies estrus; and thus mating is most likely to occur at a time when fertilization is possible. Among primates, such as the monkey, chimpanzee, and man, the display of sexual behavior is not confined to the period of ovulation but may occur at any point in the cycle (Eayrs & Glass 1962; Young 1941). The decreasing dependency on ovarian hormones has been taken as evidence that animals higher on the phylogenetic scale have through evolution become emancipated from gonadal hormone control of sexual behavior (Beach 1947b).

Females and males of some species, such as the marten and deer, are seasonal breeders. Sexual activity among seasonal breeding males is likely to be confined to periods when the females of the species are in estrus. Among nonseasonal breeding species, males do not show the cyclic display of sexual behavior characteristic of the females. Correspondingly the level of gonadal hormone in the male does not vary in any cyclic fashion.

Gonadectomy and replacement therapy

Perhaps the most compelling evidence of the importance of the gonadal hormones in the display of sexual behavior comes from studies of gonadectomy and replacement therapy.

Spaying adult females of subprimate species, with the possible exception of the rabbit, results in the complete and immediate loss of mating behavior. But the injection of estrogen or estrogen and progesterone brings the female into estrus. With appropriate timing and dosage of injected hormone, estrus thus induced may be indistinguishable from that displayed prior to spaying. In primates below man, the effects of spaying are not as conspicuous as in lower mammals, and although the effectiveness of replacement therapy has not been as carefully studied, reports suggest that the dramatic change in behavior observed in lower mammals is not seen in primates. The effects of ovariectomy on human sexual behavior are not fully known, but some sexual behavior is known to persist following the operation. In general, the less conspicuous are the changes in behavior at estrus, the less effective ovariectomy appears to be in eliminating receptivity.

Castration of the adult male does not result in complete and immediate loss of sexual behavior as does spaying in the female. The first component of the mating pattern to be lost following castration is the ejaculatory response, and later intromission drops out of the repertoire. There is considerable species variation, but in many species the male may continue to mount and show pelvic thrusts months or years after castration. The decrement in performance is generally greater in rodents and lagomorphs than in carnivores (Beach 1958). Information on the effects of castration in primates is scanty. Present evidence suggests that the prepuberally castrated monkey and ape retain the capacity for erection, and they display the normal mounting pattern (Goy 1964). Little is known of effects of castration on human sexual behavior.

Just as there is a gradual diminution of sexual behavior following castration, restoration of the complete pattern of sexual behavior by androgen administration is a gradual process. An increased frequency of mounting is followed by the occurrence of intromission and finally the ejaculatory response is restored. Such, at least, is the case in the lower mammals that have been studied. No quantitative studies exist on the effectiveness of androgen replacement therapy in subhuman primates, and the evidence for man is contradictory.

There is no question of the importance of sex hormones for the full expression of sexual behavior in adults of lower mammalian species; however, sex hormone specificity remains somewhat ambiguous. In either sex the heterotypic hormone is not as effective as the homotypic hormone in promoting the display of appropriate sexual behavior. In general, the character of the behavior induced by hormone injection, whether masculine or feminine, is not a property of the hormone itself but depends upon the organization of the sexual behavior mechanisms within the individual.

The role of gonadal hormones in differentiation of biological sex has been studied intensively by embryologists over many years (Burns 1961), but the role of these hormones in differentiation of sexual behavior has, until recently, received little attention.

It has now been demonstrated that in both the genetic female guinea pig and the rat, testosterone administration during the period of differentiation of the tissues that are to mediate sexual behavior, or what we have called sexual behavior mechanisms, results in permanent suppression of female behavior and in an increase in display of male behavior. In the guinea pig, this period of differentiation is completed prenatally, whereas in the rat the period extends into the early postnatal period. The behavior of genetic female adult rhesus monkeys treated prenatally with testosterone has not been studied, but observation of behavior during the first year of life showed a masculinization of patterns of play behavior and indications of an increase in the frequency of display of male sexual behavior (Phoenix et al. 1959; Young et al. 1964).

Further evidence indicating the significance of testosterone in differentiation of tissues mediating the pattern of sexual behavior is provided by the response of rats castrated the day of birth and injected as adults with estrogen and progesterone. These males displayed the complete female estrous response with a lordosis frequency comparable to that of females. Such was not the case when rats were castrated at ten days of age or later.

In adulthood the role of testosterone is to activate pre-established patterns of sexual behavior. During the prenatal or early postnatal period the role of testosterone is organizational, in the sense that it differentiates in both genetic males and females a prepotency for the male pattern of sexual behavior (Young et al. 1964).

The gonadal hormones are major determinants of sexual behavior both as activators and organizers of patterns of sexual behavior. The nongonadal hormones play a less direct role in control of sexual behavior. The pituitary, for example, derives its significance for behavior primarily because it controls gonadal action. Sexual behavior can be induced in hypophysectomized animals by injection of gonadal hormones, but in otherwise intact animals pituitary dysfunction is associated with abnormal reproductive function and sexual behavior.

Thyroid activity

Broad species and individual differences exist with respect to the importance of thyroid activity. The thyroid is not crucial to normal sexual behavior in the male guinea pig or rat. The thyroidectomized female rat may mate successfully, but the spayed, thyroidectomized guinea pig exhibits reduced responsiveness to injected estradiol. In the bull, thyroidectomy abolishes all sexual behavior. Present evidence emphasizes species differences, and when thyroid deficiency appears to interfere with normal sexual behavior it most likely does so by altering systemic metabolic function.

Sensory variables

The senses serve several functions vital to sexual behavior. They are essential in bringing together the male and female of a given species and in arousing and directing behavior that eventually results in copulation and reproduction. The distance receptors play a major role in locating appropriate sex partners, and all of the senses probably contribute to the arousal and performance of the behavior sequence culminating in insemination.

The effects of sensory deprivation in sexual behavior have been studied to a limited extent in the rat, rabbit, and cat. Beach (1951) has concluded that sexual arousal in the male does not depend upon any one sense modality but upon the total pattern of sensory input from the several receptor systems. Generally the female is less dependent upon multiple stimulation for display of adequate sexual behavior. A female deprived of vision, olfaction, and audition may continue to mate successfully, but this is not true of the male.

Neural variables

The neural tissues constitute what is presumed to be one of the primary loci of gonadal hormone action controlling behavior. The site of action within the nervous system is not definitely known, but it is assumed to include both spinal cord and brain.

Spinal cord

When the cord is transected above the lumbar region in males of species ranging from rat to man, stimulation of the penis still produces erection, ejaculation, and pelvic thrusting. The majority of paraplegics can achieve erection, and some have had fertile matings. It is obvious, therefore, that at least some of the more reflexive elements of the sexual behavior pattern are organized at the spinal level. Effects of cord transection in the female are not as clear, but whatever fragmentary elements of sexual behavior survive cord transection occur independently of hormonal stimulation. In neither sex is there anything resembling the normal pattern of sexual behavior in preparations with the brain stem sectioned below the level of the hypothalamus.


Small bilateral and midventral lesions in the hypothalamus of spayed female guinea pigs may eliminate all receptive behavior and mounting despite hormonal injection, which in the spayed unoperated female induces the complete estrous response (Brookhart et al. 1940). In some females with slightly different lesions, receptive behavior is blocked but mounting behavior is unaffected. In still other cases, mounting is eliminated but receptive behavior persists (Goy & Phoenix 1963). In the male guinea pig, hypothalamic lesions may eliminate all sexual behavior. However, some males may continue to mount following the operation. In the absence of testicular atrophy, loss of sexual behavior cannot be attributed to hormonal insufficiency (Brookhart & Dey 1941; Phoenix 1961).

Hypothalamic lesions in a number of species, including the rat (Soulairac & Soulairac 1956), rabbit, cat, and ewe, have resulted in comparable loss of sexual behavior. In most of these species, although not yet established for both sexes, one area in or near the hypothalamus controls pituitary function and thus governs sexual behavior indirectly; and a second area exerts direct neural control over sexual behavior. Destruction of the former eliminates sexual behavior by interfering with pituitary-gonadal axis function and therefore with output of gonadal hormone necessary to normal sexual behavior. Destruction of the latter area eliminates sexual behavior by interfering with a primarily neural mechanism (Sawyer 1960).

Lesions in the anterior portion of the hypothalamus in spayed female rats and guinea pigs have produced persistent receptivity, or estrus. Such an operation releases the animal, to some extent at least, from gonadal hormone control of sexual behavior and thus produces a condition thought to exist normally in a number of primates, especially man. The operation may be interpreted as removing a neural area that in the normal state inhibits sexual behavior unless stimulated by female hormones (Goy & Phoenix 1963).

A number of other lines of evidence implicate the hypothalamus as the region of the brain most directly involved in the control and integration of sexual behavior. For example, hypothalamic implants of solid diethylstilbestrol dibutyrate produce in the spayed cat the full display of sexual behavior. Comparable implants in other regions of the brain or in subcutaneous tissue do not produce this effect (Harris et al. 1958).

Experiments involving electrical recording from hypothalamic areas in the cat have shown changes in potential associated with estrous reactions, and changes in electroencephalograph activity of the anterior hypothalamus have also been observed in association with mating behavior in the female rabbit (Sawyer 1960).

When an electrode is permanently implanted in the hypothalamus of a rat and so wired that pressing a bar delivers a small electrical impulse to the brain area at the electrode tip, rats will repeatedly press the bar, thus obtaining electrical stimulation to the area. Castration abolishes the bar-pressing response by the rat, but the response can be reinstated by injecting the castrate with testosterone (Olds 1958). Such a demonstration tells us little about sexual behavior but implicates the hypothalamus in behavior that is dependent on a gonadal hormone.

In a now classic experiment, Kliiver and Bucy (1939) performed bilateral temporal lobectomies on monkeys and reported a broad constellation of behavioral changes, including heightened sexual activity, following the operation. Hypersexuality was later reported following destruction of the amygdaloid complex and overlying pyriform cortex in lynxes, agoutis, cats, and monkeys. Castration of male cats resulted in elimination of hypersexuality. The lesion is interpreted as having had its effect by removing from inhibition the neuroendocrine mechanisms regulating the display of sexual behavior (Schreiner & Kling 1956).

Cerebral cortex

In lower mammals, including the cat, complete removal of the neocortex of the female does not eliminate sexual receptivity, although the quality of the response may suffer. In male rats comparable decortication eliminates all sexual behavior, and these males show no signs of arousal in the presence of an estrous female. Partial decortication in the cat may eliminate successful mating behavior because of loss of sensory-motor integration, but the operation need not interfere with sexual arousal (Beach 1951).

In primates the cortex may play a major role in integration of sexual behavior. In general, the more important learning and memory are in sexual performance, the more important the neocortex is likely to be. It is apparent that control of sexual behavior is not vested in any single brain area. The evidence suggests rather that the entire nervous system is involved in the display of the normal pattern of sexual behavior.

Experiential variables

Although it has been pointed out that differences in patterns of sexual behavior are associated with differences in genetic background, this does not preclude the importance of experiential factors in determining the particular pattern displayed by the individual. The limits to which experience can modify the behavior pattern are, however, largely determined by genetic factors. In studying the influence of experiential factors on mating behavior, we are indirectly examining the variability allowed by genetic endowment.

When males from two inbred strains of guinea pigs were separated from their mothers at the age of 25 days and reared in isolation, they did not display normal mating behavior patterns when tested as adults. For isolation to produce comparable effects in male guinea pigs of a heterogeneous stock, the animals had to be isolated beginning at 10 days of age. The experiment demonstrates the influence of genetic and experiential variables in determining patterns of sexual behavior. It should be noted that not all males even within the inbred strains showed the “isolation effect” and that the relevant variables associated with isolation have not been determined (Valenstein et al. 1955).

Isolation of the male rhesus monkey at birth produces drastic effects not only on its sexual behavior but on its entire social behavior repertoire (Harlow & Harlow 1962). However, mother-deprived infants, if given an opportunity to interact with peers, develop essentially normal patterns of sexual behavior. The manner in which peer interaction fosters development of normal sexual behavior remains to be demonstrated.

Female guinea pigs and rhesus monkeys reared in isolation also show deficiencies in reproductive behavior. Despite the bizarre social behavior and abnormal sexual behavior displayed by the isolated female rhesus, a number have mated with particularly capable males and have delivered viable young. It should be noted that the behavioral deficiency is not attributable to gonadal hormone deficiency in the isolated monkey or guinea pig.

Although the exact nature and extent of the effects of isolation on sexual behavior have not been determined for the male rat, available evidence suggests that the impairment is not as profound as it is in the monkey (Zimbardo 1958).

A wide variety of experiences are capable of modifying sexual behavior in addition to the method of rearing just discussed. What has been demonstrated is that sexual behavior, in general, is subject to modification just as other aspects of behavior can be modified by varying experience. Thus despite heritability of sexual behavior patterns, dependence on hormonal stimulation, and the spinal reflex contribution to patterns of sexual behavior, the complete pattern can be blocked, enhanced, or modified by social and situational experience. The extent to which it can be modified by experience varies primarily with the species, sex, and age of the individual.

Charles H. Phoenix

[Other relevant material may be found inGenetics; Infancy, article onthe effects of early experience; Nervous system; Psychology, article oncomparative psychology; and in the biography ofKinsey.]


Beach, Frank A. 1947a Evolutionary Changes in the Physiological Control of Mating Behavior in Mammals. Psychological Review 54:297-315.

Beach, Frank A. 1947 & A Review of Physiological and Psychological Studies of Sexual Behavior in Mammals. Physiological Review 27:240-307.

Beach, Frank A. 1949 A Cross-species Survey of Mammalian Sexual Behavior. Pages 52-78 in Paul H. Hoch and Joseph Zubin (editors), Psychosexual Development in Health and Disease. New York: Grune & Stratton.

Beach, Frank A. 1951 Instinctive Behavior: Reproductive Activities. Pages 387-434 in S. S. Stevens (editor), Handbook of Experimental Psychology. New York: Wiley.

Beach, Frank A. 1958 Evolutionary Aspects of Psycho-endocrinology. Pages 81-102 in Anne Roe and George Gaylord Simpson (editors), Behavior and Evolution. New Haven: Yale Univ. Press.

Brookhart, J. M.; and DEY, F. L. 1941 Reduction of Sexual Behavior in Male Guinea Pigs by Hypothalamic Lesions. American Journal of Physiology 133: 551-554.

Brookhart, J. M.; DEY, F. L.; and RANSON, S. W. 1940 Failure of Ovarian Hormones to Cause Mating Reactions in Spayed Guinea Pigs With Hypothalamic Lesions. Society for Experimental Biology and Medicine, Proceedings 44:61-64.

Burns, R. K. 1961 Role of Hormones in the Differentiation of Sex. Volume 1, pages 76-158 in William C. Young (editor), Sex and Internal Secretions. 3d ed. Baltimore: Williams & Wilkins.

Dilger, William C. 1962 Behavior and Genetics. Pages 35-47 in Eugene L. Bliss (editor), Roots of Behavior: Genetics, Instinct, and Socialization in Animal Behavior. New York: Harper.

Eayrs, J. T.; and Glass, A. 1962 The Ovary and Behavior. Volume 2, pages 381-433 in Solly Zuckerman (editor), The Ovary. New York: Academic Press.

Ford, Clellan S.; and Beach, Frank A. 1951 Patterns of Sexual Behavior. New York: Harper.

Goy, Robert W. 1964 Reproductive Behavior in Mammals. Pages 409-441 in Charles W. Lloyd (editor), Human Reproduction and Sexual Behavior. Philadelphia: Lea & Febiger.

Goy, Robert W.; and Jakway, Jacqueline S. 1959 The Inheritance of Patterns of Sexual Behaviour in Female Guinea Pigs. Animal Behaviour 7:142-149.

Goy, Robert W.; and Jakway, Jacqueline S. 1962 Role of Inheritance in Determination of Sexual Behavior Patterns. Pages 96-112 in Eugene L. Bliss (editor), Roots of Behavior: Genetics, Instinct, and Socialization in Animal Behavior. New York: Harper.

Goy, Robert W.; and Phoenix, Charles H. 1963 Hypothalamic Regulation of Female Sexual Behaviour: Establishment of Behavioural Oestrus in Spayed Guinea pigs Following Hypothalamic Lesions. Journal of Reproduction and Fertility 5:23-40.

Harlow, Harry F.; and Harlow, Margaret K. 1962 Social Deprivation in Monkeys. Scientific American 207, Nov.: 136-146.

Harris, G. W.; Michael, R. P.; and Scott, Patricia P. 1958 Neurological Site of Action of Stilboestrol in Eliciting Sexual Behaviour. Pages 236-251 in Ciba Foundation, Symposium on the Neurological Basis of Behaviour. Edited by G. E. W. Wolstenholme and C. M. O’Connor. Boston: Little.

Jakway, Jacqueline S. 1959 Inheritance of Patterns of Mating Behaviour in the Male Guinea Pig. Animal Behaviour 7:150-162.

Kinsey, Alfred C. et al. 1948 Sexual Behavior in the Human Male. Philadelphia: Saunders.

Kinsey, Alfred C. et al. 1953 Sexual Behavior in the Human Female. Philadelphia: Saunders.

KlÜver, H.; and Bucy, P. C. 1939 Preliminary Analysis of Functions of Temporal Lobes in Monkeys. Archives of Neurology and Psychiatry 42:979-1000.

Lloyd, Charles W. (editor) 1964 Human Reproduction and Sexual Behavior. Philadelphia: Lea & Febiger.

Olds, James 1958 Effects of Hunger and Male Sex Hormone on Self-stimulation of the Brain. Journal of Comparative and Physiological Psychology 51:320-324.

Phoenix, Charles H. 1961 Hypothalamic Regulation of Sexual Behavior in Male Guinea Pigs. Journal of Comparative and Physiological Psychology 54:72-77.

Phoenix, Charles H. et al. 1959 Organizing Action of Prenatally Administered Testosterone Propionate on the Tissues Mediating Mating Behavior in the Female Guinea Pig, Endocrinology 65:369-382.

Sawyer, Charles H. 1960 Reproductive Behavior. Volume 2, pages 1225-1240 in Handbook of Physiology. Section 1: Neurophysiology. Washington: American Physiological Society.

Schreiner, Leon; and Kling, Arthur 1956 Rhinencephalon and Behavior. American Journal of Physiology 184:486-490.

Soulairac, A.; and Soulaihac, M.-L. 1956 Effets de lésions hypothalamiques sur le comportement sexuel et le tractus génital du rat male. Annales d’endocrinologie 17:731-745.

Valenstefn, Elliot S.; Riss, Walter; and Young, William C. 1955 Experiential and Genetic Factors in the Organization of Sexual Behavior in Male Guinea Pigs. Journal of Comparative and Physiological Psychology 48:307-403.

Young, William C. 1941 Observations and Experiments on Mating Behavior in Female Mammals. Quarterly Review of Biology 16:135-156; 311-335.

Young, William C.; Goy, Robert W.; and Phoenix, Charles H. 1964 Hormones and Sexual Behavior. Science 143:212-218.

Zimbardo, Philip G. 1958 The Effects of Early Avoidance Training and Rearing Conditions Upon the Sexual Behavior of the Male Rat. Journal of Comparative and Physiological Psychology 51:764-769.


Social scientific writing on sexual behavior over the past century falls in three categories: the ecology and styles of sex as a physical act; the meanings of the act, in its direct, displaced, or sublimated form, to the individual and to society; the norms and ethical standards that govern the modes of, occasions for, and designation of participants in, the physical act; and the organizational arrangements that implement these norms and standards. It is now commonly accepted that the frequencies and techniques of sexual behavior vary widely among individuals and among religious, class, ethnic, educational, age, and sex categories in the same society (Kinsey et al. 1948; 1953); that its personal meanings are numerous and contradictory (Freud 1893-1895); and that sexual norms and ethical standards contrast sharply from culture to culture (Mantegazza 1886).

As producers of literature on sexual behavior, social scientists are easily outnumbered by physicians, psychologists, physiologists, theologians, and philosophers, among others. The present article, rather than restricting itself to studies by social scientists, will conceptualize the content of this diverse literature from a social science perspective. These materials will be examined in terms of the social meanings of sexual behavior, the way the sexual enters as a component of social roles, the social norms governing recruitment to these eroticized roles, and their articulation with networks of the noneroticized roles that constitute some major social institutions. Reference will be made to social organizational arrangements for facilitating or inhibiting enactment of these roles.

The social science perspective

Sexual behavior participates in social action by contributing a motive, or a driving force of that action. Ratzenhofer (1898), an early theorist of the sociology of sex, saw all motives of social contact as modifications of either the instinct of self-preservation or the sexual instinct. Moreover, as a component of social action, sexual behavior is also a way of expressing and conveying both individual and social meanings; that is, the meaning is not only that which an individual attributes to his act but also that which society has come to apply to the act (Thomas 1907).

In contrast to the social science perspective, the “physicalist tradition” tends to treat sex simply as energy or motive. For example, the studies of Kinsey (a zoologist) and his associates at the Indiana University Institute for Sex Research follow the tradition of “biological realism” that is, they focus on the ecology and styles of the physical act. These are by far the most popular scientific studies of sex, as measured by sales and published critical reviews. From a physicalist point of view, the presumption that all acts engaging the genitals are alternative ways of tapping the same reservoir of physiological energy justifies drawing them under the same rubric. Thus Kinsey’s treatment of alternative “outlets” places greater emphasis on the channels through which energy is released than on the social context of that energy release.

Social science analyses of sexual behavior are primarily concerned with coital acts, both homosexual and heterosexual, which involve a relationship between human beings and are thus qualitatively different from other sexual behavior. Masturbation, nocturnal emissions, and bestiality claim social science attention only insofar as they affect either the occurrence of coitus or its meaning or become the object of institutional concern. According to this view, the energy problem is interwoven with that of the social meaning of sexual acts. Freudian theory of the economy of the libido uses the notion of sublimation to designate the rechanneling of sexual energy into cultural activities by shifting the object of cathexis; thus the theory interweaves motivational and meaningful aspects.

Sexual behavior contains a nonverbal language through which an individual expresses and conveys meanings; sensory exploration and gestures outweigh words as sign and symbol vehicles. Removal of clothing beyond the degree necessary for coitus itself exemplifies the significance of tactile communication. Any sexual act may carry several social and personal meanings. For example, a morning erection may express unconscious concerns about death rather than sex (Bergler 1950).

Ford and Beach (1951) studied sexual behavior between species and across cultures, yet they granted priority to the physicalist approach by classifying sexual behavior according to modes of stimulation of the genitals. A consequence of this physical perspective is their tendency to measure the cause of sexual behavior in terms of tension reduction and to designate procreation as its principal social function. The procreative function loses its centrality for a social psychology of human sexual behavior in view of the empirical fact that less than one in a thousand human coital acts results in pregnancy and fewer are intended to do so (Foote 1954). However, for sociology, the part of sexual behavior that is, in fact, procreative becomes significant because of both the demographic consequences and the institutional and organizational arrangements that become necessary for the birth and rearing of offspring (Ellwood 1925).

The eroticized role

The higher a species is in the evolutionary scale, the less dependent is its sexual behavior on gonadal hormones and seasonal physiological changes; and in man it comes under control of the cortex of the brain (Ford & Beach 1951), thus introducing the possibility of decisional and normative control. Sexual behavior as a role component, that is, as a component of social action under decisional and normative control, distinguishes human from animal sexuality.

In human sexual behavior, the erotic element may be incidental to other motives, such as the desire for a new experience or the desire to escape drudgery (Thomas & Znaniecki [1918-1920] 1958, vol. 2, pp. 1800-1821). A social role—such as that of lover, of prostitute or client, of husband or wife, of young male delinquent or adult fellator (A. J. Reiss 1961)—in which a sexual element is embedded with other social role elements, such as economic or affective exchange or child raising, may be termed an “eroticized role.” The eroticized role is part of the whole system of roles constituting the personality, and, in a sense, it eroticizes the entire personality. This is what has been described as a transformation of sexuality into Eros (Marcuse 1955). Sexuality becomes meaningful to the actor, and may be understood by the observer, in the context of these other aspects of the personality.

An eroticized social role may be articulated with other roles in the social structure; for example, the role of the prostitute in her relation with a client may be meshed with her relatively nonerotic role in her relation with her madam or other employer. This perspective on role articulation follows Malinowski (1929), who advocated an anthropological perspective that does not treat sex as a mere physiological transaction but studies its implication for lovemaking, its function as the nucleus of the family, the spells and magic that grow up around it, its effects upon, and how it is affected by, the legal system and the economic system.

The variety of sexual meanings is, in part, dependent on other aspects of the role—that is, on other norms, affects, or meanings that define that role or other roles with which it is articulated in a personality or in a society. Kirkendall (1961) described how sex, while it may advance a deep interpersonal relation, may be exploitative as part of a casual relationship. With prostitutes, the relationship itself becomes the instrumental servant of its sexual component. Even in advancing a deep relation, however, sex may assume special meanings associated with the broader meaning of that relationship. Thus sex may be a weapon to destroy, a mode of defense, a bargaining point, or a form of self-assertion; negatively, it may be a form of self-denial (Frank 1954).

Components of an eroticized role may be malintegrated. A marriage contracted for social or economic advantages despite sexual incompatibility is a classic instance. Sexual behavior inconsistent with personal ethical standards may engender an internal struggle manifested in guilt feelings. If strains of this type become widespread, social standards may be readjusted (I. L. Reiss 1960).

Recruitment to the eroticized role

Recruitment of partners to an eroticized role is governed by considerations of personality, social norms, and societal mechanisms of control. Personality variables affect the desire and ability of an individual to seek, as well as his desirability as, a sex-role partner. Physiological or attitudinal disturbances in an individual’s response at any stage of the coital act influence recruitment to the eroticized role. In the preparatory stage, an individual must be able to respond to excitation from internal and external psychological stimulation. This requires sensory, intellectual, motor, and glandular contributions to mechanisms of arousal. An individual’s attitude toward himself—for example, as reflected in body image–affects his ability to woo another through foreplay and enter intimate relationships (Fisher & Cleveland 1958). An individual must experience impetus to penetrate or desire to be penetrated, want to continue inplay, and, at culmination, experience orgastic peristalsis of genital structures. Much depends upon ability to tolerate regression in the service of the ego (Fried 1960). Disturbances of sexuality are commonly classified in terms of frigidity, low potency, preference for masturbation or for sexual perversion, hypersexuality, and lack of sexual interest (Eisenstein 1956).

Special personality groups vary in interest in eroticized roles. Most of the erotic impulse of idiots is directed toward themselves and the remainder to objects of the same sex (Potter 1927). Alcoholics lack sex interest, because, it has been argued, of repressed homosexuality (Levine 1955). The aged tend to lose motivation for coitus (Armstrong 1963).

The evaluation of personal characteristics in recruitment is relative to the persons involved; that is, an individual not desirable to one partner may be desirable to another. For example, males generally prefer shorter females, and females generally prefer taller males. Women who want men near their own height, it has been argued, are expressing a need for ascendancy. Similarly, males who seek taller women are said to be seeking to possess the forbidden parent figure (Beigel 1954).

Norms of recruitment

Social norms prescribe and proscribe which social types may recruit one another. Havelock Ellis (1910), for example, wrote that in choosing a mate we tend to seek parity of racial and anthropological characteristics together with disparity of secondary sexual characteristics and complementarity of psychic characteristics. Under control of instinct, animals rarely depart from parity of species and disparity of secondary sexual characteristics. Mongrelization may be induced among domesticated animals but is extremely rare among wild animals despite opportunities. Man can, and sometimes does, engage in relations that the more constitutionally directed animals would refuse. Man may disregard species parity and practice bestiality, disregard secondary sexual disparity and practice homosexuality, and disregard psychic complementarity and establish neurotic sex relations. On the other hand, man may refuse relations because of sexual style, the status of the partner, or the beauty of the partner—a parity irrelevant to the animal. For man, the power of the social over biological determinants extends to the very definition of gender. Money et al. (1957) compared the relative influences of the chromosomal sex, gonadal sex, hormonal sex, internal accessory reproductive structures, and external genital morphology on the gender role actually assumed by hermaphrodites, and they found social sex assignment to be the most reliable prognosticator of gender role.

Rules for recruitment to coital relations in general parallel, but do not completely overlap with, those governing marital recruitment. Laws of exogamy define the inner social limits for recruitment of a marital partner. These are extensions of the incest taboo, which is a nearly universal taboo, with exceptions only in such cases as the Egyptian royal family, where brother-sister marriage was permitted, or among Azande nobles, where father-daughter marriage has been permitted. Laws of endogamy define the outer social limits. These norms have been compared by Westermarck (1889) to the “law of similarity” that keeps animals from pairing across species. In human relations, however, endogamy controls potentially fertile relations within the same species.

A number of individuals are virtually precluded from recruitment to eroticized heterosexual roles. This is the case with institutionalized populations in prisons or mental hospitals. Likewise, occupational conditions may separate individuals from the opposite sex. Hobos, being homeless, have less access to women, and thus they tend to frequent the lowest status prostitutes and to practice homosexuality.

Social control of eroticized role recruitment contributes to the maintenance of the social structure; if society lacked such controls, random mating could disrupt the familial and economic lines of stratification of a society. Moreover, social control of recruitment may enable one social group to dominate another through its power to allocate sexual occasions and facilities. It may also be an internally integrative device, as it is among celibate monastics. According to Ross (1920), such control is necessary to manage a smoldering antagonism that exists between the sexes as groups.

Sexual communication in recruitment

Communication employed for recruitment to the eroticized role parallels the communication of arousal in the sex act. Token arousal announces sexual availability. Erotic communication relies heavily on movement, such as the inviting movements of the dance and the arousing tactile movements during coitus. Odors founded in the natural capryl odors and supplemented at the human level by perfumes act upon the deeper levels of consciousness (Bloch 1907). Cosmetic means of communicating sexual interest include painting and tattooing as well as sexual mutilation. Socially stylized ceremonies of fertility or initiation may announce availability and effect arousal. A few rites, such as subincision, announce sexual availability but reduce sexual enjoyment or arousal potential for the subject (Allen 1949; Westermarck 1889).

Verbal communication supplements sensory communication in sexual behavior. Speech is so closely identified with sexuality that in some cultures it is forbidden or restricted between persons who are socially precluded from engaging in sexual relations (Baker 1949). The more intimate a sexual matter, the less the likelihood of communication about it to nonerotic role partners.

In many societies, socially structured mores about erotic communication bar young people from arousal and consummatory opportunities. Sexual norms may be learned inferentially by the young or by new social groups, such as immigrants, by observation of behaviors lacking primary erotic meaning, such as forms of dress and of association between the sexes (Klausner 1964). Classified newspaper advertising and a specialized journalism directed to erotica have long been used for the recruitment of both heterosexual and homosexual partners.

Deviant subgroups develop private signaling systems to cue a potential partner without revealing themselves to noninitiate spectators. For example, homosexuals may wear peculiar clothing (the definition of which changes from time to time), intersperse their speech with a special argot, or perform certain gestures. They may position themselves to receive such signals by frequenting known haunts.

Social control of sexual behavior

Society not only exerts control over the recruitment of role partners but also attempts to control the types of sexual behaviors enacted and the occasions for their enactment. Both informal and formal social controls are found. Attitudinal disapproval of individuals’ erotic behaviors, which is perhaps the commonest form of control, may be expressed as verbal chastisement, gossip, scandal, humor, or mockery. The Trobriand Islanders, for example, control sexual style by jesting about those who use the “white man’s position” for intercourse. A quasi-formal control mechanism is exemplified among the Mondurucu by the gang rape to force a recalcitrant female to submit to male authority (Murphy 1959).

Norms also define those behaviors to which it is legitimate to recruit. Early studies of sexual behavior treated cross-cultural comparisons descriptively, presenting various behaviors, sexual art, and technical devices as curiosities (Moll 1912; Krafft-Ebing 1886; Bloch 1907). These studies did, however, open the way to the study of sexual norms by giving information about criteria for recruitment to eroticized roles in other cultures. Correlations between sexual behavior and the overarching cultural ethos allow inference about the types of sexual behaviors to which recruitment is approved in various cultural situations; that is, recruitment norms vary with the ethos. For example, Sorokin (1956) argued that sexual asceticism is associated with “idealistic societies” and libertarianism with “sensate societies.” Taylor (1953) classified historical periods in Western society as predominantly “matristic” or “patristic.” In patristic periods, dominated by a father religion and exemplified by England during the first millennium of the Christian era, people have a horror of homosexuality, sex is generally restricted, and rape and incest increase. In matristic periods, dominated by a mother religion and exemplified by England after the first millennium, people have a horror of incest and are permissive regarding sex.

Social control may be exerted by concrete organizational arrangements. Organizations that facilitate recruitment to the sex role also maintain procedures, such as the “sex game,” for controlling recruitment and defining types of permitted alliances (e.g., see Cressey 1932). The Nazis, while promoting a policy of sexual abstinence, provided youth camps in which sexual relations took place. Monasteries may debar their inmates from sexual opportunities. Child marriage commits the relationship before the erotic drive appears. The sexes may be physically and, therefore, socially isolated, as in a harem or school. Chaperonage is a way of supervising the young during their meetings. Systems of peer group pressures, such as those exerted by campus sororities or fraternities, encourage socially approved love relationships by subjecting them to ratings by age mates. These controls are more stringent among the upper classes, who have more to lose by a breakdown in the stratification system (Goode 1959).

Legal control takes the form of laws against homosexuality, adultery, sodomy, or pederasty—that is, against behaviors that violate rules governing recruitment to the eroticized role. In some jurisdictions, the style of coitus is legally as well as informally directed. In England, the Wolfenden Committee on homosexual offenses and prostitution gathered intelligence about sexual behavior from social scientists and psychiatrists as a basis for legislating on sexual recruitment. They concluded that it is the duty of the law not to concern itself with immorality as such but to confine itself to activities that offend against public order and decency. Thus, they recommended that homosexual relationships between adults by mutual consent and in privacy should not fall within the province of criminal law (Great Britain . . . 1957). Federal legislation in the United States, such as the Mann Act (the so-called “white slave” traffic act) of 1910, is given force through policing activities of the Immigration and Naturalization Service, the Federal Bureau of Investigation, and local police departments. Throughout Anglo-American history, the methods of legal control of sex expression have varied much more than has the accepted doctrine of sexual morality (May 1931; Mueller 1961).

Sex education is a form of noncoercive, though formal, control. The medical profession in its efforts to control venereal disease, as well as religious groups in their efforts to maintain their own sexual norms, have been instrumental in founding organizations such as the American Society of Sanitary and Moral Prophylaxis in 1905 and the American Social Hygiene Association in 1912, now the American Social Health Association. This latter association uses education to combat venereal disease, commercialized prostitution, and other conditions associated with sex delinquency among young people.

Resistance to social control

Protest against efforts to enforce prevailing sexual mores and to control recruitment to the eroticized role is expressed in individual sexual deviance, the formation of groups of individuals with special sexual needs, and movements to change the prevailing mores. The sex offender uses illegitimate means, such as violence or disregard for the consent of the partner, to recruit to coital or noncoital sexual behaviors, or he recruits without regard to rules defining permitted sexual partners. Generally, these individuals are undersexed rather than oversexed. Freudian theory suggests that some sex offenders may be compensating for feelings of bodily damage or phallic inadequacy, or suffering from castration anxiety (Hammer 1957). Most sex offenders are of average intelligence. Those arrested for incest, however, tend to be older men of subnormal intelligence and antisocial personalities. Apprehended distributors of obscene literature tend to be older, sexually inhibited men of average intelligence who are not hostile. Women are rarely arrested for sexual offenses in American society, except for violating norms regarding public solicitation. The role relations in primarily nonsexual delinquent groups may become eroticized, involving, for instance, group masturbation.

Organizations of individuals whose physical status makes it difficult for them to recruit or be recruited to an eroticized role, such as dwarfs or the blind, may oppose prevailing sexual mores, as may those who, though not necessarily handicapped or abnormal in a clinical sense, have sexual requirements that are contrary to the norms. These groups provide evaluational support, facilitate recruitment to the role, and insulate members against the sexual mores of the larger society (Hooker 1956).

Social movements concerned with sex have sought revision of sexual standards, of the laws supporting them, and of the means of enforcing them. Corruption in systems of police regulation of prostitution led to the formation of the International Federation for the Abolition of State Regulation of Vice in 1875. The World League for Sexual Reform on a Scientific Basis, perhaps the most extensive such group, was an outgrowth of the Institute for Sexual Science founded by Magnus Hirschfeld in Germany in 1929. It attempted to abolish antisexual taboos, establish a philosophical basis for sexual freedom, and provide education in the form of lectures and clinical advice. The institute was closed by the Nazis, and the world movement collapsed a few years later. Wilhelm Reich, who criticized Hirschfeld for attempting sexual reform within the framework of capitalist society, established the Socialist Society for Sexual Advice and Study in Vienna in 1928 to “defend the rights of children and young people to their natural love.”

Sex and social institutions

The nonerotic components in an eroticized role may link it with a pattern of roles constituting an institution. Historically, eroticized roles have been part of the fabric of economic, political, religious, familial, and other institutions. At certain times, a particular institution, such as the church in Christian society, may exclude the erotic component from its roles and influence its place in the roles of another institution, especially the family.

The family

Social movements to reform sexual mores ordinarily protest the family’s monopoly of the legitimate eroticized role. The eroticized role is universally legitimated in the family, and in modern Western society it is legitimated only in the family. However, actual behavior is at variance with this official standard. In the Kinsey samples, 50 per cent of the females and 85 per cent of the males with a high school education reported that they had experienced premarital coitus. Nearly half of these females had experienced coitus with a man other than their fiancè. Moreover, by the age of forty nearly 26 per cent of the females and 50 per cent of the males had engaged in extramarital coitus.

Increasing opportunities for premarital and extramarital coitus accompany the greater mixing of the sexes in the educational, occupational, and social worlds. Legal control of premarital intercourse in American society is generally enforced more strictly in the case of younger participants. Adultery is grounds for divorce in all states. To the extent that coitus takes place outside of marriage the procreative aspect declines in significance. For example, Malinowski (1929) thought that the Trobriand Islanders, who have very permissive sexual mores, did not appreciate the connection between coitus and pregnancy, although it is more likely that, while aware of the connection, they were not much interested in it. Discussions of contraception reveal some tensions aroused concerning nonprocreative coital meanings, since contraceptives are not simply designed to prevent conception but to do so while coitus continues.

Sex and religion

Sexual behavior has been interwoven with religion from earliest times. Religion’s general concern with fertility involves it in the procreative aspect of sex, and this concern may even become pervasive, as, for example, in the Indian lingam cults. Furthermore, sexual energy itself may play a role in religious worship. Scott (1941) concluded that phallic worship originated in the pleasure associated with coitus and not in any clearly conceived notion that intercourse would produce children. Sex as worshipful, expressive communication with a supraindividual force is illustrated by the activities of temple prostitutes in Babylon. Krafft-Ebing (1886) argued that religion and sex could replace one another, since both display a similar scope and quality of excitement at their peaks and both can degenerate under pathological conditions into cruelty. Religious mystical roles may contain an erotic compound in both the Western and Eastern traditions. On the other hand, the cases of the sacerdotal celibate, the religious hermit, or the penitent abstainer illustrate religious exclusion of the sexual component.

Religious considerations affect recruitment to eroticized roles in other institutions. Thus, rules of exogamy and endogamy are generally supported by religious sanctions. Religious sanctification of marriage or the withholding of that sanctification affects the recruitment of coital partners. Moreover, religion may contribute to the isolation of the sexes by separating them in worship and by excluding women from sacred precincts.

Religious devotion is negatively associated with participation in nonreligious eroticized roles in the United States. The frequencies for all sexual outlets, except for marital coitus, decrease with increasing devoutness (Kinsey et al. 1948; 1953). This association seems to hold in France as well (Institut Francais . . . 1960).

Sex and the polity

The relation of sexual behavior to the nature of the state and politics has long been of popular concern. The literature on court ladies and on espionage has described the impact of eroticized roles upon political roles. Sorokin (1956) asserted that sexual overindulgence debilitates the entire society. Krafft-Ebing (1886) saw a two-directional relation, in which, on the one hand, sexual excess undermines the props of society and, on the other hand, the collapse of society produces sexual aberrations.

Utopian schemes have included recommendations for control of eroticized roles. Unwin (1940) proposed that, in his new society, Hopousia, there be two kinds of marriage: alpha marriage, which would be monogamous and preceded by prenuptial continence, for those who would achieve, create, and lead, and beta marriage, which may be terminated at will and would not require prenuptial continence, for those who do not aspire to social position. Unwin’s thesis is grounded in the concept of energy sublimation. In John Humphrey Noyes’ Oneida community, a cooperative colony founded in 1848, extramarital “coitus interruptus” was permitted, on the ground that sex is a way of communicating the meaning of intimate friendship.

The acquisition of political power by men usually implies an access to more women. For example, in Bedouin society, a tribal chief is more likely to be polygamous and to retain concubines than is a less powerful member of the tribe. In Europe, royalty as well as the economically powerful have been able to keep mistresses. Conquest has generally implied access to the women of the conquered society, either by rape, by capture, or through the woman’s desire to consort with the powerful.

Sex and the economy

Sex has a place in the economy both as a marketable commodity and as a component affecting the marketability of nonsexual commodities. Thus economic value can be derived from sexual value. For example, prostitution involves an economic-sexual exchange of this character; the economic exchange ramifies beyond the prostitute and her client to include the pimp who recruits the client, the madam who supplies facilities for the act, and, perhaps, the police who protect all of them from the law. In the case of marriage, the economic-sexual exchange is embedded in a wider set of family and community relations. Sexual role performance consumes facilities, goods, and services, and thus industries emerge around the provision of facilities for recruitment to the eroticized role and for enactment of that role. These range from the provision of meeting places, such as dance halls; the publishing of pornographic, scientific, artistic, and guidance manuals; the supply of pharmaceutical and mechanical contraceptives; and the letting of places, such as a motel room or brothel, for consummating the sex act.

Sexual value may be derivative from economic value as well. For example, sexual jealousy, rather than being an innate emotion, may be related to views of women as property. Jealousy is then an anger aroused when property rights are violated. A norm of premarital chastity may also be related to marketability (Nemeček 1958).

Reich (1930), using syncretistic Marxian-Freudian “sex-economic” concepts, described an interplay between sexual and economic norms. He reasoned that since psychic structure is created by social structure and the core of psychic structure is the sexual function, sexual function is controlled by social structure. Social structure is constituted by the “modes of production,” including the interpersonal relations of production. These relations affect sexual needs, among others. They also influence concepts of life, morals, and philosophy—that is, ideology–and so influence sex norms. Classes standing in different relation to the “modes of production” will differ in their sexual needs and norms. The dominant political minority imposes its ideology in general, and its sexual ideology in particular, on a population. According to this view, the subordinate population thus has sex needs derived from its own class position while being subject to the sexual ideology of another class. It should be noted that Kinsey’s findings do not support Reich; Kinsey and others (1953) found little relation between the parental occupational class and the incidence of various sexual behaviors.

Research methods

Methods of social and psychological research developed in other areas have been used to ascertain frequencies of sexual behaviors as well as social and individual meanings associated with that behavior. Questionnaires (Christensen & Carpenter 1962) and interviews (Kinsey et al. 1948) have both been used to elicit such information. Protecting the anonymity of the respondent and establishing the scientific legitimacy of the interviewer are salient bases for rapport during an interview. To maximize the validity of responses, the interviewer must avoid making the respondent feel guilt, anxiety, or defensiveness. Direct questions should be asked with no defensiveness or apology on the part of the interviewer, and terminology should be adapted to the social-class level of the respondent. The inquiry should begin with matters that do not provoke anxiety, such as age, and then proceed to items which are sexual but for which the individual does not feel responsible; thus the interviewer would ask about spontaneous orgasms before asking about masturbation. The fact that anxiety associated with each type of behavior differs among culture or class groups and between the sexes affects the ordering of the questions. For example, homosexuality is less frightening than masturbation for females in American society, while the reverse is true for males; thus, in interviewing females, questions on homosexuality precede those on masturbation, and the reverse order is used for males. As a practical procedure one might request the age at first orgasm, then ask for its source and, if this is not too anxiety-provoking, trace this source through the years. By asking when he first experienced each type of behavior, the burden of denial is placed upon the interviewee. Kinsey’s interviewers severely censured any subject suspected of giving a dishonest history.

Kinsey correlated the frequencies of each source of outlet—masturbation; premarital, extramarital, and marital coitus; homosexuality; and animal contacts—with demographic factors such as national origin, marital status, age, education, occupational class, rural or urban residence, religious affiliation, and religious devoutness, among others. Intrapsychic meanings of sexual behavior have been studied by correlating projective test responses with the frequencies of behaviors. The Rorschach test has been used to reveal body image (Fisher & Cleveland 1958), the Thematic Apperception Test to uncover feelings of guilt associated with sex (Leiman & Epstein 1961), and a combination of the Thematic Apperception Test and the House-Tree-Person Test to obtain data about castration feelings or feelings of phallic inadequacy for psychoanalytic interpretation (Hammer 1957). Anthropological field methods, employing observation and informant interviews, have been used to gain information about both the occurrence of behaviors and their meanings (Malinowski 1929). Observational methods supply data on systems of communication and interaction that are necessary for an analysis of the social organization of sexuality (A. J. Reiss 1961). The Human Relations Area Files, containing field reports on many societies, have been used to exploit the advantages of both field observation and correlation methods (Murdock 1949). Attitude questionnaires have been used to obtain information on ideal prescriptive norms, as has content analysis of newspapers, radio, and plays (A. Ellis 1954).

The fact that the various sexual behaviors may be ordered along dimensions of greater or lesser intimacy or greater or lesser acceptability has been exploited to construct scales. Podell and Perkins (1957), for example, developed a Guttman scale for ordering types of sexual experience, and Ira Reiss (1964) developed a refined Guttman scale for a study of standards of sexual permissiveness.

Samuel Z. Klausner

[See alsoFamily; Marriage; Prostitution; Sexual behavior, articles onsexual deviationand onhomosexuality; and the biographies ofEllis; Freud; Malinowski; Reich; Westermarck.]


Primary literary materials as sources of data may be located in several extensive bibliographies: Hayn & Gutendorf 1875 lists references and abstracts of erotica in German; Reade 1936 lists and abstracts English material; and Gay 1861 includes a similar collection of French materials. Ellis & Abarbanel 1961, containing articles and short bibliographies on most topics of interest, is a good introductory resource for secondary materials.

Allen, Clifford 1949 A Study of Sexual Mutilations. International Journal of Sexology 2:158-161.

Armstrong, Eunice B. 1963 The Possibility of Sexual Happiness in Old Age. Pages 131-137 in Society for the Scientific Study of Sex, Advances in Sex Research. Edited by Hugo G. Beigel. New York: Harper.

Baker, Sidney J. 1949 Speech and Sexual Taboos. International Journal of Sexology 3:98-101.

Beigel, Hugo G. 1954 Body Height in Mate Selection. Journal of Social Psychology 39:257-268.

Bergler, Edmund 1950 Morning Erections. International Journal of Sexology 3:141-144.

Bloch, Iwan (1907) 1937 The Sexual Life of Our Time. New York: Falstaff. → First published as Das Sexualleben unserer Zeit in seinen Beziehungen zur modernen Kultur.

Christensen, Harold T.; and Carpenter, George R. 1962 Timing Patterns in the Development of Sexual Intimacy: An Attitudinal Report on Three Modern Western Societies. Marriage and Family Living 24: 30-35.

Cressey, Paul G. 1932 The Taxi Dance Hall. Univ. of Chicago Press.

Eisenstein, Victor W. (editor) 1956 Neurotic Interaction in Marriage. New York: Basic Books.

Ellis, Albert (1954)1962 The American Sexual Tragedy. 2d ed. New York: Stuart.

Ellis, Albert; and Abarbanel, Albert (editors) 1961 The Encyclopedia of Sexual Behavior. 2 vols. New York: Hawthorn.

Ellis, Havelock (1910) 1936 Studies in the Psychology of Sex. Volume 4: Sex in Relation to Society. New York: Random House.

Ellwood, Charles A. 1925 The Psychology of Human Society. New York: Appleton.

Fisher, Seymour; and Cleveland, Sidney E. 1958 Body Image Boundaries and Sexual Behavior. Journal of Psychology 45:207-211.

Foote, Nelson N. 1954 Sex as Play. Social Problems 1:159-163.

Ford, Clellan S.; and Beach, Frank A. 1951 Patterns of Sexual Behavior. New York: Harper.

Frank, Lawrence K. 1954 The Psychocultural Approach in Sex Research. Social Problems 1:133-139.

Freud, Sigmund (1893-1895) 1955 The Standard Edition of the Complete Psychological Works of Sigmund Freud. Volume 2: Studies on Hysteria, by Josef Breuer and Sigmund Freud. London: Hogarth. → First published in German.

Fried, Edrita 1960 The Ego in Love and Sexuality. New York: Grune.

Gay, Jules (1861) 1894-1900 Bibliographie des ouvrages relatifs à I’amour, aux femmes, au mariage .... 4th ed. 4 vols. Paris: Becour.

Goode, William J. 1959 The Theoretical Importance of Love. American Sociological Review 24:38-47.

Great Britain, Committee on Homosexual Offenses and Prostitution (1957) 1963 The Wolfenden Report: Report of the Committee on Homosexual Offenses and Prostitution. New York: Stein & Day.

Hammer, Emmanuel F. 1957 A Psychoanalytic Hypothesis Concerning Sex Offenders. Journal of Clinical and Experimental Psychopathology 18:177-184.

Hayn, Hugo; and Gutendorf, Alfred N. (1875) 1912-1929 Bibliotheca germanorum erotica et curiosa. 9 vols., 3d ed. Munich: Muller. → Hayn was the sole author of the first edition, published as Bibliotheca germanorum erotica.

Hooker, Evelyn 1956 A Preliminary Analysis of Group Behavior of Homosexuals. Journal of Psychology 42: 217-225.

Institut Franqais D’Opinion Publique (1960) 1961 Patterns of Sex and Love: A Study of the French Woman and Her Morals. New York: Crown. → First published in French.

Kinsey, Alfred C. et al. 1948 Sexual Behavior in the Human Male. Philadelphia: Saunders.

Kinsey, Alfred C. et al. 1953 Sexual Behavior in the Human Female. Philadelphia: Saunders.

Kirkendall, Lester A. 1961 Premarital Intercourse and Interpersonal Relationships. New York: Julian.

Klausner, Samuel Z. 1964 Inferential Visibility and Sex Norms in the Middle East. Journal of Social Psychology 63:1-29.

Krafft-Ebing, Richard F. von (1886) 1965 Psychopathia sexualis. New York: Putnam. → First unexpurgated edition in English; original in Latin.

Leiman, Alan H.; and Epstein, Seymour 1961 Thematic Sexual Responses as Related to Sexual Drive and Guilt. Journal of Abnormal and Social Psychology 63:169-175.

Levine, Jacob 1955 The Sexual Adjustment of Alcoholics: A Clinical Study of a Selected Sample. Quarterly Journal of Studies on Alcohol 16:675-680.

Malinowski, Bronislaw (1929)1962 The Sexual Life of Savages in North-western Melanesia: An Ethnographic Account of Courtship, Marriage, and Family Life Among the Natives of the Trobriand Islands, British New Guinea. New York: Harcourt.

Mantegazza, Paolo (1886) 1935 The Sexual Relations of Mankind. New York: Eugenics Publishing. → First published in Italian.

Marcuse, Herbert 1955 Eros and Civilization: A Philosophical Inquiry Into Freud. Boston: Beacon.

May, Geoffrey 1931 Social Control of Sex Expression. New York: Morrow.

Mead, Margaret 1949 Male and Female: A Study of the Sexes in a Changing World. London: Gollancz; New York: Morrow.

Moll, Albert (1912) 1921 Handbuch der Sexualwissenschaften. 2d ed. Leipzig: Vogel.

Money, John; Hampson, J. G.; and Hampson, J. L. 1957 Imprinting and the Establishment of Gender Role. A.M.A. Archives of Neurology and Psychiatry 77:333-336.

Mueller, Gerhard 1961 Legal Regulation of Sexual Conduct. Legal Almanac Series, No. 9. Dobbs Ferry, N.Y.: Oceana.

Murdoch, George P. 1949 The Social Regulation of Sexual Behavior. Pages 256-266 in Paul Hoch and Joseph Zubin (editors), Psychosexual Development in Health and Disease. New York: Grune.

Murphy, Robert F. 1959 Social Structure and Sex Antagonism. Southwestern Journal of Anthropology 15: 89-98.

NemeČek, Ottokar 1958 Virginity: Prenuptial Rites and Rituals. New York: Philosophical Library.

Podell, Lawrence; and Perkins, John C. 1957 A Guttman Scale for Sexual Experience: A Methodological Note. Journal of Abnormal and Social Psychology 54:420-422.

Potter, Howard W. 1927 An Introductory Study of the Erotic Behavior of Idiots. Journal of Nervous and Mental Disease 65:497-507.

Ratzenhofer, Gustav 1898 Die sociologische Erkenntnis. Leipzig: Brockhaus.

Reade, Rolf S. 1936 Registrum librorum eroticorum .... 2 vols. London: Privately printed.

Reich, Wilhelm (1930) 1945 The Sexual Revolution. 3d ed. New York: Orgone Institute Press. → First published as Die Sexualität im Kulturkampf.

Reiss, Albert J. Jr. 1961 The Social Integration of Queers and Peers. Social Problems 9:102-120.

Reiss, Ira L. 1960 Premarital Sexual Standards in America. Glencoe, III.: Free Press.

Reiss, Ira L. 1964 The Scaling of Premarital Sexual Permissiveness. Journal of Marriage and the Family 26:188-198.

Ross, Edward A. (1920) 1938 Principles of Sociology. 3d ed. New York: Century.

Scott, George R. (1941) 1952 Phallic Worship. New York: Mental Health Press.

Sorokin, Pitirim A. 1956 The American Sex Revolution. Boston: Sargent.

Taylor, Gordon R. 1953 Sex in History. London: Thames & Hudson.

Thomas, W. I. 1907 Sex and Society: Studies in the Social Psychology of Sex. Univ. of Chicago Press.

Thomas, W. I.; and Znaniecki, Florian (1918-1920) 1958 The Polish Peasant in Europe and America. 2 vols., 2d ed. New York: Dover. → See especially Volume 2, pages 1800-1821, “Sexual Immorality of Girls.”

Unwin, Joseph D. 1940 Hopousia: Or, the Sexual and Economic Foundations of a New Society. London: Allen & Unwin.

Westermarck, Edward A. (1889)1921 The History of Human Marriage. 3 vols., 5th ed. London: Macmillan.


One premise of psychoanalytic doctrine is that the general developmental process of childhood includes psychosexual development. It may be taken as a general premise of all growth, physical and psychological, that at any one stage of development the organism’s current growth status and future potential is determined by (1) what it began with, (2) what has been added or subtracted in the interim, perhaps irreversibly, and (3) the growth-facilitating or growth-impeding environment that it may meet. Thus, at the moment of conception, growth status and potential is determined by both the quantitative and qualitative aspects of the genetic code of information, supplied half by the mother, half by the father, and by the intrauterine environment in which the fertilized egg exists. The intrauterine environment may prove hostile mechanically, metabolically, or because of the invasion of foreign organisms, viruses for example. The embryo itself may be its own enemy, as when a particular developmental phase or system fails, with subsequent deleterious effects on another phase or system. For example, if the testes differentiate imperfectly in a genetic male, then the fetal testicular secretions fail, with the result that the anlagen of the external genital morphology develop as either completely or partially female, depending on the degree of testicular failure. Contrarily, in the case of a genetic female, if the fetal adrenal cortex functions erroneously, so as to produce an excess of androgen in the place of expected cortisone, then the anlagen of the external genitalia will masculinize in their development, even to the point that the genital tubercle becomes a fully formed penis instead of a clitoris.

Principle of the critical period

Both of the foregoing examples of an error in embryonic sexual development illustrate the principle of the critical period, that is to say, that period in the development of some system or function when the organism passes from a neutral or undifferentiated stage to a differentiated one. It is during this critical period that development is vulnerable to interference and deflection, which, if they occur, will leave a permanent residual in structure or function. The further development of the organism will then in some degree be determined by, or influenced by, the outcome of the critical period.

Effects of atypical early experiences

The concept of the critical period applies not only to morphologic and prenatal development, but to behavioral, psychological, and postnatal development as well. In behavioral science, the European animal ethologists were the first to observe that patterns of behavior, including sexual behavior, can be experimentally changed during a critical, developmental learning period, and thenceforth this deviance becomes relatively fixed. In other words, particular atypical experiences may interfere with or deform the genetic, hormonal, or neural norms of behavioral development otherwise expected.

The justly famous experiments of Harlow (see Money 1965a, chapter 7) on the macaque monkey are a case in point. The experimental monkeys showed that opportunity to play with age-mates in early childhood was a developmental prerequisite to the achievement of normal copulatory ability in adolescence. Depriving the infants of childhood play, combined with depriving them of interaction with their mothers by rearing them with dummy mothers made of wire covered with a piece of towel, had particularly adverse effects. Such doubly deprived animals were unable to adapt to the motions of mating and to effect sexual intercourse even when paired with mates selected for experience and gentleness. Those few females in whom pregnancy was finally achieved were incompetent to mother their own young. They neglected and injured them, sometimes cruelly. Other aspects of their behavior, in general, also appeared to be grossly disordered.

Another illustrative experiment is that of Birch (1956). Female rats were raised in isolation with a rubber collar or ruff around their necks, which deprived them of the experience of licking themselves or others. They were finally released from their collars to deliver their first litter. Instead of licking the young as they were born, they ate most of them. The surviving 5 per cent were retardedly and improperly retrieved and badly suckled; some were eaten after being carried to the nest. Three survivors of a large litter died of starvation. Their mother had shepherded them under her chin every time they approached her belly to search for a nipple.

Experiments in population density with rats have demonstrated yet another way in which behavioral experience, namely crowding, may induce aberrant sexual and social behavior (Calhoun 1962), affecting even the unborn fetuses and their future behavior, presumably via the maternal endocrine system (Keeley 1962).

The proper understanding of psychosexual development and of erroneous development in human beings requires focusing attention on experience, behavior, and social interaction during the formative years of childhood. In man, perhaps even more so than in lower species, peculiarities and special features of experience during critical periods may change the expected genetic, hormonal, or neural norms of development.

Hermaphroditism—an example

Probably the most dramatic illustration of this last proposition is to be found in the psychology of hermaphroditism (Money, Hampson, & Hampson 1957; Money 1961). Human hermaphroditism, for which there are several different etiologies, is conventionally defined in terms of a discrepancy between the gonads (ovaries, testes, or, rarely, ovotestes) and the morphology of the external genitalia. Sometimes the discrepancy may be total, as indicated at the beginning of this article, namely, when the gonads are testicular and the external organs are perfectly female (the syndrome of testicular feminization) or when the external organs are perfectly male except for an empty scrotum, the gonads being ovaries and in proper ovarian position (a rare variant of the hyperadrenocortical syndrome of female hermaphroditism). Most commonly in hermaphroditism, however, the discrepancy between gonads and external organs is incomplete, by reason of the fact that the external organs are themselves imperfectly differentiated as either male or female. In this incompleted state of differentiation, the external sex organs of either sex look remarkably similar: the penis with a urinary gutter instead of a tube and with its orifice near the scrotum can easily be confused with an enlarged clitoris, and the incompletely fused scrotum may pass for labia that have improperly begun to fuse. Because confusion is possible and because medical decisions may sometimes differ, it may so happen that two individuals of identical genetic, gonadal, and hormonal diagnosis are assigned to different sexes.

Effects of sex assignment

The expected outcome of such sex assignment is that psychosexual differentiation will proceed to take place congruously with assignment and rearing. Ideally, the external organs will have been repaired surgically as early as possible, so as to conform to the assigned sex. The visible, kinesthetic, and tactile body image is a potent fact in a child’s own awareness of his or her gender identity, and the visible appearance of sex is, of course, the ultimate criterion for the gender-role expectancies imposed by most of the people who are significant in a child’s life.

Even with delayed genital repair, it has proved possible for psychosexual differentiation to take place in keeping with assigned sex when genetic, gonadal, and hormonal sex are at variance. In congruous-looking sex organs militate against psychosexual differentiation appropriate to sex assignment, especially when the incongruity is reinforced by some other source of ambiguity such as parental lack of conviction as to the way they should be rearing their child or open doubt expressed by age-mates. In some instances, doubt may develop in the child’s own mind without any clear retrospective indication of its origin.

Ideally, for psychosexual differentiation to be brought to maturity at puberty, assigned sex should not be at variance with hormonal sex and the secondary sexual changes of the body that are hormonally controlled. This ideal can be met in modern endocrinology. Yet, there are occasional untreated cases; among them are cases of hormonal virilization in girls with the hyperadrenocortical syndrome, which is remarkable in view of the intensity of virilization and its early onset before the usual age of puberty. These cases show that the persistence of a feminine psychosexual differentiation is possible despite the hormonal contradiction and its severe handicap to proper adolescent social maturation.

Crucial variables

In psychosexual differentiation, the crucial variables, which may in hermaphroditism be independent of one another and may be overridden by the experiential effects of assignment and rearing, are five: (1) chromosomal sex, as determined by actual chromosomal count, (2) gonadal sex, (3) hormonal sex, (4) internal accessory morphologic sex, and (5) external morphologic sex. Psychosexual differentiation itself transcends not only all five of these variables but also the assigned sex in the syndrome of transvestism with transsexualism, which has, therefore, been treated, and not without reasonable success, by surgical, hormonal, and social sex reassignment (Benjamin 1964; Pauly 1965).

To have found rare clinical cases in which the five physical variables can be overridden is not, of course, to have proved that the physical variables do not under other circumstances contribute to, or help determine, psychosexual differentiation. What has been proved is not that psychosexual differentiation is always and exclusively a matter of behavioral and experiential life history but that experiences determined by rearing are far more potent than might otherwise have been expected. Harlow’s investigations on the macaque demonstrated, on the one hand, the extraordinary power and permanence of the social-experience factor in preventing psychosexual maturity; but they also led him, on the other hand, to conclude that sex differences in normal macaque behavior antedate childhood experience and learning.

Harlow observed (1962) that the young male monkeys made more threats toward other monkeys, both male and female, than did the young females.The females’ threats, moreover, were reserved primarily for other young females. The young females retreated more often than the males, specifically by adopting the female sexual posture. The young males initiated more play contacts, with playmates of either sex, than did the females; and the males had a monopoly on rough-and-tumble play. With increasing age, the males showed increasing frequency of the male mounting position in their copulatory play. They showed little grooming behavior, which, in adults, is more a characteristic of female than male sexual behavior.

The application of Harlow’s method of direct observations and systematic experiments on monkeys is restricted in human beings, as compared with clinical case study, by reason of the usual ethical vetoes on experimental manipulation of the lives of human beings and by the customary taboos imposed on sexual play and the expression of sexual interests in childhood. Systematic experimental attempts to study the development of masculinity and femininity have confined themselves to the secondary data of questionnaires or interviews or to the oblique data of free association, projective tests, and play. Although not thoroughly consistent or definitive, the findings show that psychosexual differentiation varies with age and in some degree is related to family and class differences in child-rearing practices and the emotional quality of the relationship between the child and each parent.

Sex reassignment

That there is a critical period in psychosexual differentiation, and that it is early in life, is suggested by clinical evidence, again from hermaphroditism, namely in instances of a change of the sex of assignment (Money, Hampson, & Hampson 1956). It is not difficult for a child under a year of age to adjust to a sex reassignment, provided the parents can also adjust. But the older the child—for example, if he is past the age of learning to talk and is becoming acquainted with the verbal and other signals of gender difference—and the further that psychosexual differentiation has progressed without ambiguity, the progressively less likely it is that an imposed sex reassignment will be successfully assimilated. It thus appears that the establishment of a psychosexual identity and gender role roughly parallels the learning of the native language. The critical period, in which the basic essentials are laid down, is ended probably by the kindergarten age. The remaining years of childhood are years of consolidation until the hormonal changes of adolescence bring psychosexual differentiation to its completed expression in both the subjective and the behavioral aspects of falling in love, courtship, mating, and parenthood.


Hormonal puberty is temporarily delayed sometimes, and sometimes it never occurs, except for the intervention of hormonal replacement therapy. Without the advent of hormonal puberty, psychosexual differentiation remains to some degree in a state of infantilism. The exact degree varies from individual to individual, for the psychosexual maturation of the teen-ager is equated to a certain extent with social maturation, as is well demonstrated in cases of physical sexual precocity where the body becomes sexually mature as early as age six but the social maturation remains consistent with chronological age and/or degree of social exposure. A few hormonally prepubertal teen-agers are able to overcome the handicap of a sexually unappealing, prepubertal body morphology and to keep up socially with their age-mates. But the task is difficult and the competition harsh, so that the majority lag in social maturation. They retreat, as it were, to a younger age group, waiting for the day when their bodies no longer tell a lie about their ages. If this waiting is too protracted, there may be a permanent scar in social development.

Those late pubertal developers who succeed in not lagging in social maturity may even succeed also in appearing to have reached psychosexual maturity, in the sense that they have sexual relations, copulate to the partner’s satisfaction, and perhaps get married. But their achievement is, in fact, a simulation, insofar as their proprioceptive erotic sensations and cognitional eroticism in general are not the same as they will be after hormonal treatment has begun. A male, for example, whatever peak of erotic experience he may reach, does not have an orgasm as he will later experience it. He does not have sex dreams incorporating orgasm, wet or dry; he does not have masturbation imagery, or much of an urge to masturbate; and he does not respond with the same frequency or intensity to erotic perceptual stimuli, visual and narrative, as he will after hormonal puberty is induced. Probably he is incapable of experiencing an episode of that very intense concatenation of feeling so little attended to scientifically—namely, falling in love.

Failure to develop full normal maturity of psychosexual differentiation after hormonal puberty, natural or induced, may represent one of four types of failure in the growth process, regardless of etiology. One is a simple arrestment of development that had been otherwise normal up to that point. The second is a regression from a once more advanced level. The third is the introduction of a fault or error in development. The fourth is a continued persistence of developmental error from an earlier period. In psychosexual differentiation, a persistent error may have had its genesis in faulty differentiation at the outset of childhood or perhaps in regression occurring in a later period of childhood.

It is probable that a large proportion of psychosexual pathology in evidence at adolescence does not have its initial origins in postpuberty but represents the persistence of faulty psychosexual differentiation earlier. The fault manifests itself in full only under the influence of hormonal maturation. Conspicuous effeminacy in boys, for example, has been noted from as early as the age of three (Green & Money 1961) and seen to progress from later childhood through puberty to chronic, overt homosexuality (unpublished data). Theories of the life history dynamics of homosexuality have recently been comprehensively reviewed by Pritchard (1962) and by Bieber and his associates (see Society of Medical Psychoanalysts 1962).

Whatever the type or timing of developmental error, the responsible or contributing etiologic agent may be, singly or in combination, genetic, gestational, metabolic, nutritional, toxic-infective, traumatic, behavioral-experiential, social-interactional, or otherwise. The cold scientific fact of the matter is that astonishingly little is known of the etiology of psychosexual pathologies. Knowledge of their phenomenology and natural history is greater (Money 1963c), although it is far from complete.

Brain mechanisms

Current advances are being made in knowledge regarding brain mechanisms mediating sexual behavior. Brain ablation was the earliest experimental technique for such investigation, gains and losses in sexual behavior being related to damaged or removed neural centers and pathways. A newer technique is that of direct electrical stimulation of neural centers and pathways by means of micro-electrodes implanted in the brain.

The neurophysiology of sex is best understood, even if somewhat simplified, in terms of what MacLean (1962) has delineated as man’s three brains: the upper mammalian brain or new cortex; the lower mammalian brain or old cortex, also known as the limbic system; and the reptilian brain, comprising the brainstem reticular system, the midbrain, and the basal ganglia. The limbic cortex attaches to the reptilian brain like a pair of headphones covered by the helmet of the neocortex. In the reptilian brain, the centers of the hypothalamus lie in close proximity to the pituitary gland. The hypothalamic-pituitary axis is of fundamental importance in sex, for it is here that the nervous system and the endocrine system meet—here that neural messages become neurohumoral, and neurohumoral ones become hormonal. It is here that the hypothalamus regulates cyclic functioning of the pituitary gonadotropic hormones, which in turn regulate the hormonal functioning of the gonads. Here then, indirectly, lie some of the dictates of behavior related to the effecting of pregnancy and parenthood. The limbic system also collaborates.

Electrode implantation studies

In the ringlike arrangement of structures on each side of the brain that form the limbic system, there is in the upper forward part the septum and below it the amygdala. MacLean, exploring the representation of sexual function in the limbic system of the squirrel monkey by means of implanted electrodes, found that partial sexual functions could be directly activated by stimulation of the septal area. In the adjacent area of the amygdala chewing and salivation were activated, with partial erection occurring after many seconds of stimulation or as a rebound phenomenon after the termination of stimulation. MacLean conjectured that the topographical representation of the head end and the rear end of the body in such close proximity bears some relationship to the intimate interplay of behavior in the oral and sexual spheres. In lower animals, this interplay is in evidence in sexual smelling and in the delivery of the young. In human beings it appears in ordinary love play. Exaggerated and distorted, it appears also in the pathologies of anal-oral and urethral–oral stimulation and in stimulation from smelling, smearing, or ingesting feces or urine.

The proximity of representation of all the phylogenetically primeval action systems and their related nonverbal feeling or emotion in the old brain may help to explain other pathologies of sex, such as the association of sexual stimulation with mutilation and attack in masochism and sadism. In some of his experiments, MacLean obtained a relation between seminal discharge and genital scratching, which, unless experimentally interrupted, would become severely self-mutilative (MacLean, Dua, & Denniston 1962).

Neurohormonal studies

Another new direction in the neurophysiology of sex lies in neurohormonal research (reviewed by Money 1965b). One approach is to inject a sex hormone into the body and then study its effect on sexual centers in the limbic system and in the hypothalamus and adjacent midbrain nuclei. In one group of rabbit studies, Sawyer and his co-workers (Kawakami & Sawyer 1959) demonstrated the effect of subcutaneously injected sex hormone on electroencephalogram activity, registered from implanted electrodes, relative both to local vaginal stimulation and to sexual response. Olds (1958, p. 683) and Critchlaw, working with male rats experimentally prepared for brain self-stimulation, found that with some electrode placements and certain low electric shock, self-stimulation responses disappeared almost completely after castration and reappeared, progressively, after replacement therapy with testosterone propionate in oil.

A second approach in neurohormonal research is to implant hormones directly into sexual centers in and around the hypothalamus. Fisher (1956) used the technique of implanting fluid hormone through a microcannula. By introducing minute amounts of testosterone into the preoptic area of male rats, he obtained rather bizarre combinations of mating and maternal behavior, including nest building and persistent retrieving and grooming of foster litters of young. Harris, Michael, and Scott (1958) used the improved technique of implanting microamounts of solid hormone and were able to induce a state of sustained sexual receptivity in cats, in the absence of genitopelvic estrus. Michael (1961; 1962) used radioactive estrogen and traced its site of action in the brain by making autoradiographs of the sliced brain. The autoradiographic technique further established (Glascock & Michael 1962) that individual neurons in the hypothalamus take up radioactive estrogen that has been given by intramuscular injection.

These various implantation neurohormonal experiments have shown that the sex hormones are not clearly sex specific, but often paradoxical in their neurobehavioral effect. Other chemicals may be found capable of producing their effects. The ultimate possible application of this new breakthrough in dealing with psychosexual pathology cannot be foreseen at the present time.

A third approach in neurohormonal research is to alter the neural organization of later sexual behavior by injecting hormones into the pregnant mother or newborn animal (guinea pig and rat). The findings of several investigators converge, tentatively indicating that, dependent on species differences in gestational age at birth, there is a critical period around the time of birth when androgen has an organizing effect on neural centers in the hypothalamus. This effect abolishes the pituitary’s ability subsequently to produce gonadotropins cyclically, as in the female; and it allows mating pattern elements that are chiefly masculine, although they are shared by both sexes, to predominate in childhood.

Some dimensions of deviance

Androgen as an organizer

Androgen is thus implicated as an active organizer of the sexual control system, both in terms of reproductive cycles and sexual behavior. At a critical period earlier in embryonic life, androgen is an active organizer substance without which the internal anlagen of the female reproductive anatomy persist and the anlagen of the external organs feminize (Jost 1958). Nature’s primary impulse, in other words, is to make a female. One conjectures a relationship here with the greater vulnerability of the male to disease and death, which would create an unequal sex ratio except for the fact that 106 males are born to 100 females. One conjectures also a connection with the fact that psychosexual disorders have a higher incidence in men than in women, with some of the psycho-sexual anomalies, like fetishism, voyeurism, and exhibitionism, being to all intents and purposes unheard of in females.

Sex differences in arousal patterns

The higher incidence of psychosexual pathologies in the male may be related to a sex difference in erotic arousal patterns. This difference is manifested in the greater capacity of the male to be aroused erotically and genitopelvically, and to be ready for immediate release, by pictorial and narrative imagery. Such stimuli arouse the female sentimentally to want her lover or husband; but she is dependent on direct tactile stimulation, more than the male, for the arousal of genitopelvic eroticism.

There is some evidence in the psychology of hermaphroditism that the male arousal pattern may be neurally organized under the influence of androgen in fetal life. It will be for further research to determine whether the development of psychosexual pathology in the male may be rooted in part in imperfect or faulty neurohormonal organization at this early period. One thing already quite clear in the evidence of hermaphroditism is that a masculine type of arousal pattern is perfectly compatible with entirely feminine cognitive content and imagery of eroticism that parallels the sex of assignment.

Arousal stimuli

Perceptual stimulants to sexual arousal and the reproductive cycle vary with the species. The relation between light stimulation and the release of pineal hormones is being newly investigated in the rat. Olfactory stimuli also arethe subject of important new experiments (Parkes & Bruce 1961), which show, after copulation in the mouse, that odors released by the male regulate the establishment of pregnancy in the female. Acuity of the sense of smell in human females (Money 1965b) is regulated by estrogen and varies with the menstrual cycle. Its relation to other cyclic sexual phenomena or to sexual disorder has not been investigated. Odor may be the agent responsible, in the macaque monkey, for fluctuation in male sexual behavior in rhythm with the female menstrual cycle, as described by Michael and Herbert (1963).

In addition to benefiting from the new neurophysiological techniques, contemporary sex research has also benefited from the new techniques in genetics. These techniques have enabled researchers to establish the presence (female) or absence (male) of the sex-chromatin mass (Barr body) in the nucleus of the body’s cells and to photograph, measure, and count the actual chromosomes in cells. The relevance of the new findings to the theory of psychosexual pathology has been chiefly negative: gross chromosome defects are not demonstrable in the majority of psychosexual disorders. A partial exception may exist in the case of Klinefelter’s syndrome. Affected individuals are phenotypic males with an extra X chromosome. There appears to be an unusually high incidence of psychopathology, including sexual disorders, that develops in association with the syndrome (Money 1963a). Among the sexual disorders are homosexuality and transvestism (Money & Pollitt 1964). Otherwise genetics cannot be directly implicated in psychosexual ambiguity (Money 1963b).

Disorders of sexual psychology express themselves in full in adolescence after the advent of hormonal puberty brings erotic functioning to maturity. It is probable that the majority of such disorders have their initial origins in faulty psychosexual differentiation persisting from early childhood. Later error or regression of differentiation are also possible, as is simple arrestment in psychosexual growth.

Faults, errors, regression, and arrest of development have, on the one hand, a chronologic origin and, on the other hand, an etiologic origin. The chronologic origin may fall within a critical period, with resultant immutability or refractoriness to change. The etiologic origin may be single or multiple. An etiologic factor may by itself represent only a potential vulnerability to psychosexual disorder, materializing only when in combination with another factor or factors. Thus a genetic vulnerability may materialize only when the behavioral environment conspires appropriately. Etiologic considerations include genetic, gestational, metabolic, nutritional, toxic-infective, traumatic, behavioral-experiential, and social-interactional factors.

John Money

[Other relevant material may be found inAffection; Developmental psychology; Genetics; Identity, psychosocial; Infancy, article Onthe effects of early experience; Nervous system; Psychoanalysis; Socialization.]


Benjamin, H. 1964 Nature and Management of Transsexualism: With a Report on Thirty-one Operated Cases. Western Journal of Surgery, Obstetrics and Gynecology 72:105-111.

Birch, H. G. 1956 Sources of Order in the Maternal Behavior of Animals. American Journal of Orthopsychiatry 26:279-284.

Calhoun, John B. 1962 Population Density and Social Pathology. Scientific American 206, Feb.: 139-148.

Fisher, Alan E. 1956 Maternal and Sexual Behavior Induced by Intracranial Chemical Stimulation. Science 124:228-229.

Glascock, R. F.; and Michael, R. P. 1962 The Localization of Oestrogen in a Neurological System in the Brain of the Female Cat. Journal of Physiology 163: 38P-39P.

Green, Richard; and Money, John 1961 Effeminacy in Prepubertal Boys: Summary of Eleven Cases and Recommendations for Case Management. Pediatrics 27:286-291.

Harlow, Harry F. 1962 The Heterosexual Affectional System in Monkeys. American Psychologist 17:1–9.

Harris, G. W.; Michael, R. P.; and Scott, P. P. 1958 Neurological Site of Action of Stilboestrol in Eliciting Sexual Behavior. Pages 236-254 in Ciba Foundation, Symposium on the Neurological Basis of Behavior. Edited by G. E. W. Wolstenholme and C. M. O’Connor. Boston: Little. → Includes two pages of discussion.

Jost, Alfred 1958 Embryonic Sexual Differentiation: Morphology, Physiology, Abnormalities. Pages 15-45 in Howard W. Jones and William W. Scott (editors), Hermaphroditism, Genital Anomalies and Related Endocrine Disorders. Baltimore: Williams & Wilkins.

Kawakami, M.; and Sawyer, Charles H. 1959 Neuroendocrine Correlates of Changes in Brain Activity Thresholds by Sex Steroids and Pituitary Hormones. Endocrinology 65:652-668.

Keeley, Kim 1962 Prenatal Influence on Behavior of Offspring of Crowded Mice. Science 135:44-45.

Maclean, Paul D. 1962 New Findings Relevant to the Evolution of Psychosexual Functions of the Brain. Journal of Nervous and Mental Disease 135:289-301.

Maclean, Paul D.; Dua, S.; and Denniston, R. H. 1962 A Mapping of Cerebral Structures Involved in Seminal Discharge and Genital Scratching. American Neurological Association, Transactions 87:136-139.

Michael, Richard P. 1961 An Investigation of the Sensitivity of Circumscribed Neurological Areas to Hormonal Stimulation by Means of the Application of Oestrogens Directly to the Brain of the Cat. Pages 465-480 in International Neurochemical Symposium, Fourth, Varenna, Italy, 1960, Regional Neurochemistry . . . Proceedings. Oxford: Pergamon Press.

Michael, Richard P. 1962 Estrogen-sensitive Neurons and Sexual Behavior in Female Cats. Science 136: 322-323.

Michael, Richard P.; and Herbert, J. 1963 Menstrual Cycle Influences Grooming Behavior and Sexual Activity in the Rhesus Monkey. Science 140:500-501.

Money, John 1961 Hermaphroditism. Volume 1, pages 472-484 in Albert Ellis and Albert Abarbanel (editors), Encyclopedia of Sexual Behavior. New York: Hawthorn.

Money, John 1963a Cytogenetic and Psychosexual Incongruities With a Note on Space-Form Blindness. American Journal of Psychiatry 119:820-827.

Money, John 1963b Factors in the Genesis of Homosexuality. Pages 19-43 in Conference on Community Mental Health Research, Fourth, Washington University, St. Louis, 1962, Determinants of Human Sexual Behavior: Proceedings. Springfield, III.: Thomas.

Money, John 1963c Psychosexual Development in Man. Volume 5, pages 1678-1709 in Encyclopedia of Mental Health. New York: Watts.

Money, John (editor) 1965a Sex Research: New Developments. New York: Holt. → See especially Chapter 7, “The Effect of Rearing Conditions on Behavior,” by Harry F. Harlow and M. K. Harlow.

Money, John 1965b Influence of Hormones of Sexual Behavior. Annual Review of Medicine 16:67-82.

Money, John; Hampson, J. G.; and Hampson, J. L. 1956 Sexual Incongruities and Psychopathology: The Evidence of Human Hermaphroditism. Johns Hopkins Hospital Bulletin 98:43-57.

Money, John; Hampson, J. G.; and Hampson, J. L. 1957 Imprinting and the Establishment of Gender Role. A.M.A. Archives of Neurology and Psychiatry 77:333-336.

Money, John; and Pollitt, Ernesto 1964 Cytogenetic and Psychosexual Ambiguity: Klinefelter’s Syndrome and Transvestism Compared. Archives of General Psychiatry 11:589-595.

Olds, James 1958 Self-stimulation Experiments and Differentiated Reward Systems. Pages 671-687 in Herbert H. Jasper et al. (editors), Reticular Formation of the Brain. Boston: Little.

Parkes, A. S.; and Bruce, H. M. 1961 Olfactory Stimuli in Mammalian Reproduction. Science 134:1049-1054.

Pauly, Irab. 1965 Male Psychosexual Inversion: Transsexualism. Archives of General Psychiatry 13:172-181.

Pritchard, Michael 1962 Homosexuality and Genetic Sex. Journal of Mental Science 108:616-623.

Society of Medical Psychoanalysts 1962 Homosexuality: A Psychoanalytic Study, by Irving Bieber et al. New York: Basic Books.


Deviance in general, and sexual deviance in particular, presents the scientist with extremely difficult problems of definition and analysis. Sexuality is, perhaps more than any other aspect of human behavior, mingled with moral imperatives, conscious fantasy, and unconscious desires, all of which combine to trouble even most scientific observers (Trilling 1950; Rieff 1959). The recognized necessity (not to mention utility) of sexual behavior makes the problem of establishing cutoff points beyond which the interaction of actors, acts, and contexts may be denned as deviant even more obscure than it is in other areas of behavior. The positive functions of the sexual impulse for society have indeed resulted in advocacy, by small minorities to be sure, of behavior that is widely conceived of as immoral, antisocial, or immature.

In a discussion of sexual deviance it would be useful to consider cross-cultural data. However, there is a minimum of scientifically comparable data about sexual behavior available, even from European countries (see Kinsey et al. 1953, pp. 95-96 for a list of such research), and a systematic attempt to discuss sexual deviance in non-Western cultures, where only the barest outline of information about sexual behavior is available, would be presumptuous. A useful compilation of the available cross-cultural information has been made by Clellan S. Ford and Frank A. Beach (1951). In this article the cultural focus is mainly the United States; other industrialized Western societies are treated with less precision.

Sexual prohibitions and their violation

Sexuality has the curious distinction of being the only biological drive that has been regulated in nearly all its physical manifestations. Although these prohibitions are often attributed to the prudery of the nineteenth century, it is clear that they are of considerably more ancient origin. All the sexual acts that are currently legislated against in the criminal law owe their stigma to the Judaeo-Christian religious tradition. These very early prohibitions remain with us today, when, in a very different context and serving other functions, they still operate to shape training for sexual behavior, attitudes toward this behavior in both the self and in others, and the institutional structures that are meant to control it.

Although it is possible to find some situation in which any act might be described as deviant, it is very difficult to approximate the universal condemnation of deviant sexuality in any other area of behavior. Particularly impressive is the discrepancy between the system of sexual taboos, as expressed in both the law and individual attitudes, and the corresponding systems of overt behavior. From the survey research on sexual behavior in the United States (see Kinsey et al. 1953, pp. 94-95), it is apparent that violations of widely held taboos against many forms of sexual behavior are frequent. There is a considerable bias toward greater male participation in deviant behavior than female, but both sexes include a relatively large number of taboo violators. In addition, for the relatively short period for which data are available these rates of participation have remained remarkably stable, indicating that they are outgrowths of normal social operation rather than the consequence of upheaval or disorder.

There is evidence that although there is a wide distribution of these different forms of sexual behavior, attitudes toward them remain relatively repressive (see Wheeler 1960 and I. Reiss 1960 for attitudes toward premarital coitus). Repressive attitudes can be held even by those who habitually participate in deviant sexual activities, as has been shown by a study of homosexually experienced males (Gebhard et al. 1965, pp. 638-639). In addition, there is some evidence that these attitudes are situation-specific and change over the individual’s life cycle; for instance, Robert Bell and Jack Buerkle (1961) found that the attitudes of mothers toward their daughters’ having premarital coitus varied with the age of the mother, the age of the daughter, and other factors affecting mother–daughter relationships.

Sexual learning as a source of deviance

Given the strong taboos against most of its manifestations, it is very difficult to determine why so much deviant behavior occurs. However, it is important to point out that although the taboos are strongly expressed, the sanctions are not as strongly invoked. At the social-psychological level the major question seems to be “What are the reasons for the attachment to the prohibitory rules, coupled with the failure to apply sanctions?”

Part of this is probably due to the sheer strength of the biological drive. It is important not to assume that all of the variance in sexual behavior is attributable to social and psychological factors, since there is still a general lack of data on the interaction between these elements and the biological substrate. Some of the differences between the overt sexual behavior of males and females may be due to basic hormonal and other biological characteristics as well as to differential socialization practices (Kinsey et al. 1953, chapters 14-18; Money 1961; Young, Goy, & Phoenix 1964).

The supreme importance of early childhood experience for sexual development—an insight that we owe to the early work of Freud (1905)—has been amply demonstrated by the work of such researchers as John Money (1961), John and Joan Hampson (1961), Harry Harlow (1959), and René Spitz (1949). In the first few years of life, interactions between parents and children, predominantly at the nonverbal level, build a basis of sexual capacities on which later verbal learning concerning normative standards of behavior about these capacities is overlaid. Furthermore, the majority of adult-child interaction having impact on the learning of sexual behavior is not conceived of by the adult as having any relevance to the ultimate sexual adjustment of the child. Such adult-child interaction is permitted or even encouraged, provided that its latent sexual content is not explicitly labeled or identified for what it is; as soon as it is so labeled and acknowledged it falls within the purview of one prohibition or another. Because of these learning conditions the child grows up in what has been called a nonlabeled or mislabeled sexual environment (Sears, Maccoby, & Levin 1957, pp. 176-217).

It is the ambiguity of early sexual instruction that seems to be important. The setting of the capacities and directions of sexual behavior occurs at an unrecognized and unconscious level, whereas the conscious attitude instruction not only takes place sporadically but frequently is unrelated to the development of the behavioral systems that the attitudes purport to control. One of the functions of language is that of helping the child to order and to control the world; apparently, what happens in the formation of sexual attitudes is that words are not supplied to the child to enable him to label and come to grips with his feelings and desires. In such a context the sexual zone is left to the imagery and fantasy often noted in childhood (and adult) sexuality.

Examination of the social structures that control sexual learning after childhood makes apparent the imprecision and vagueness of the connections that exist between behavior on the one hand, and attitudes toward such behavior, on the other. A certain amount of sexual display and exploration occurs during the prepubertal years among members of both sexes, but the significant element is that unless this activity is noted and defined negatively by adults or unless the behavior takes place between an experienced person and a child, there is little impact on the child. In addition, there is not necessarily any organization of sexual learning as a result of these experiences. A child who has had prepubertal experience similar to coitus may be unaware of the connection between this behavior and other sexual activities.

During puberty the significance of the genital sexual function becomes more apparent, but young people who become involved in sexual behavior do so in considerable ignorance. In the culture of adolescence the folklore of sexual experience is rife with error and misperception. Males and females get together sexually often as the result of accident, and frequently they do so for nonsexual motives. Because of the strength of the sexual drive and the ambiguous circumstances under which its control is learned, much adult sexual behavior takes place under conditions of mutual misunderstanding. Although there is a transfer of learning from one sexual experience to another, there is rarely a successful dialogue between male and female about such matters of mutual sexual interest. The bulk of sexual communication is gestural and, between persons with some mutual experience, often ritual in character.

Forms of sexual deviance

Deviance is traditionally conceived of as being antithetical to the values of the larger social system, although it is often thought to have value because punishment of it reinforces conformity. Many sociologists have described deviance as being learned in a socially well-defined initiatory system (for example, the predelinquent subculture or even lower-class society in toto) and maintained in a supportive social setting (for example, the drug-addict community). A primary weakness of such theories of “differential association” (Sutherland & Cressey [1924] 1960, pp. 74-81) or differential identification in the explanation of deviance is that they require an interactional system in which the actor learns the values and techniques of the deviant life. For most sex offenders, and for nearly all classes of sexual behavior that are described as deviant, a prior socially structured learning system apparently does not exist. In addition, the notion that deviant behavior must be maintained through subcultural supportive systems is difficult to apply to the problem of sexual deviance, although in some types of sexual deviance such systems do appear.

Sexual deviance seems to be of three general types. The first is situation-specific deviance that is normatively disvalued (or at least not valued positively) but which nonetheless serves generally useful purposes and is so ubiquitous that sanctions are applied to only a minority of those involved. This type of deviance involves no articulated system of social roles. The second type of sexual deviance consists of those forms that have supportive social organizations either for the purpose of gathering sexual partners or clientele together, or for social support in deviance, or for both purposes. Finally, there is a large body of deviance (pedophilia, incest, and so on), which is condemned and against which strong sanctions are invoked, that does not seem to be learned in recognizable deviant social structures and that does not have a socially organized performance system.

Functional deviance

Masturbation and premarital coitus are primary examples of behavior that is generally denned as deviant but that does not involve the application of sanctions unless the perpetrators are, by some mischance, apprehended or there is some untoward consequence, such as pregnancy or venereal disease. These forms of behavior may be described as “wrong but normal.” There is considerable class variation in the incidence and frequency of masturbation and premarital coitus, with the former important among middle-class males and the latter among lower-class males. Solitary masturbation is not against the law; premarital coitus may be, especially if either party is legally a minor, but even so prosecutions are comparatively rare.

The perceived utility of masturbation as a male adolescent sexual outlet with few social consequences in comparison to coitus may account for the decreasing stigma attached to it. It seems to be related to other behavior defined by social class and characterizes males who assume middle-class occupational roles (Kinsey et al. 1948, chapter 10). Premarital coitus is culturally valued for males especially within the pro-masculine lower-class culture, accounting for its prevalence and the lack of invoked sanctions (I. Reiss 1960; Whyte 1943). Both masturbation and coitus are class-linked and seem to be the most appropriate responses to the impact of the sexual drive after puberty in the male. For females premarital coitus is connected with impending marriage and a love relationship; thus, the behavior is defined as proper in a situation-specific way. Masturbation in females seems to have no class linkages at all. In general these activities have a smaller incidence and a lower frequency among females than among males, possibly because of both hormonal differences and differences in socialization.

Deviant subcultures

The second type of deviant sexual behavior is that which involves supportive social structures either for the purpose of entry into the system of behavior or for the maintenance of the individual in the behavior. This type of behavior comprises homosexuality and prostitution. It is only female prostitution that has all the elements of a deviant subculture similar to that of drug addicts or juvenile delinquents (Davis 1961; Kinsey et al. 1948, chapter 10). In this situation the stigma for the sexual behavior is attached to the prostitute, and the legal sanctions for the behavior are leveled primarily against her, secondarily against the abetting persons (pimps, madams), and exceedingly rarely against the customers or against agents of law enforcement who may condone her activities. Prostitution develops organized forms primarily because of its commercial aspect and because of the character of the enforcement program that is designed to control it.

For the most part, recruitment of individuals for homosexual experience does not occur through a social system organized for that purpose. The majority of persons who come to conceive of themselves as homosexual do so on the basis of readying experiences early in life that are linked to secondary reinforcements in puberty. Unfortunately, studies of the etiology of homosexuality have been most unsatisfactory in their attempts to locate the crucial early experiences that are salient (Money 1963). Some etiological studies have focused on biological and especially genetic factors, sometimes mediated through hormonal anomalies. But even though there have been some studies of twins (Kallman 1952; 1963) and some work has been done on hormonal factors (Kinsey 1941), the most persuasive discussions of early factors in the development of a homosexual commitment have been couched in the psychoanalytic tradition or in certain variants of it (Society … 1962; Ovesey 1954). The psychoanalytic tradition locates the origins of homosexuality in pathological versions of parent-child relations in which the mother provokes her son’s sexual interest and in consequence burdens him with an anxiety about incest with which he cannot deal. As a result his relations with females are blighted because of his fears of paternal revenge for his incest, which remains forever unconsummated. Other explanations in this tradition are less metaphorical in their language but still focus upon seductive mothers, detached fathers, lack of a male figure in the home, and the like. Unfortunately the bulk of the populations who have been the source of the data in these reports have been examined in the course of psychotherapy or other treatment, creating a situation in which pathology and homosexuality are inextricably confounded.

Psychological research into homosexuality has also maintained a determined concern with the etiological question and with developing instrumentation that might successfully distinguish between the homosexual and the heterosexual on the basis of either manifest items or latent items that are presumed to be indicators of homosexual commitments (see, for instance, the psychometric indices of masculinity and femininity used by Brown 1958; Hooker 1957; 1958; Dean & Richardson 1964; Freund 1963).

During the pubescent years there is a good deal of minor homosexual experimentation that accounts for nearly all of the homosexual activity of the predominantly heterosexual portion of the population. Thus, it can be predicted that about one in three of all males born in the United States will at some time have a homosexual experience to the point of orgasm; however, about 85 per cent of those having such experience will do so only in their adolescent years. Among females the proportion who ever have a homosexual experience will be somewhat smaller, but once again most of them will confine the experience to adolescence (Kinsey et al. 1948; 1953). The factors that distinguish those who go on to adult homosexual commitments from those who experience homosexual acts only as transient parts of adolescent development are still unknown. However, it is known that males often develop strong sexual interests in other males without knowing that there is such a thing as homosexuality. (This process occurs more frequently in specialized sexual interests such as transvestism and fetishism.)

The realization of being homosexual occurs in a variety of ways: sometimes through processes of introspection, sometimes through falling in with other homosexuals who act seductively, and sometimes through sexual contacts with other males who finally become defined as homosexual. It is after the point of conscious admission that one is a homosexual—an admission that ranges from the traumatic to the delighted—that the search for sexual partners begins. This search usually requires entry into the world of homosexual bars, development of the argot of homosexual life, and gradual learning of the modes of approach and retreat that are related to the satisfaction of sexual needs. Study of this zone of the homosexual’s adjustment is beginning to result in a literature that suggests that adult homosexual commitments are not reenactments of early pathological relations between parents and children. These studies have begun to outline the processes of adult socialization that are involved in the development of the homosexual commitment. There now also exists a literature on particular aspects of the homosexual life (Hooker’1966; Leznoff & Westly 1956; Achilles 1967).

The extant literature is nearly exclusively concerned with the male homosexual; the literature on the lesbian is both scanty and scientifically inadequate. In general, discussions of the lesbian tend to focus on the similarities between male and female homosexuality and to see one as the mirror image of the other—that is, to see male homosexuals as imitations of females and female homosexuals as imitations of males. A more satisfactory view, however, is to see female homosexuality as continuous with normal female sexuality (Simon & Gagnon 1967; Ovesey 1956).

Not all homosexuals are regular members of the homosexual community and many of them never involve themselves in this social milieu at all. It is difficult to determine whether or not a more intensive or a more detached relationship to the community is more helpful in the homosexual’s personal and social adjustment. The maintenance of female homosexuality occurs more often outside the community context than does homosexuality among males (Simon & Gagnon 1967).

The development of homosexual contacts among prisoners in penal institutions is more like an articulated system of introduction to homosexual experience than is homosexual development in the free community. Homosexual behavior in prisons ranges from the careful seduction, in adult male institutions, of a weaker male by an aggressive male (involving money and protection from the latter in exchange for the sexual cooperation of the former, and often leading to “marriage”) to the “extended family” systems observed in juvenile female institutions and the homosexual “bringing out” in female penitentiaries (Lindner 1948; Sykes 1958; Halleck & Hersko 1962; Lambert! 1963; Ward & Kassebaum 1965; Giallombardo 1966).

The antisocial deviant

The third type of deviant sexual acts to be considered are those most specifically prosecuted by law-enforcement agencies; they involve aggression, youth, close kin relations, and public disturbance (a typology suggested both in Wheeler 1960 and in Indiana University 1965; see pp. 1-13 for the classification of sex offenders). Most of these acts are committed by individuals whose behavior is the result of needs developed outside any subcultural or deviant training ground composed of other persons oriented toward the same behavior. The origins of heterosexual pedophilia, for example, seem to lie in the incapacity of the male to establish any meaningful relationship with an adult female or in many cases a failure to find any sociosexual outlet at any level of psychological involvement. The pedophile resorts to children because the difference in age gives him a sense of power over them as well as the hope that he will meet with an uncritical reception. There is no social support for the act either before or after its occurrence. The individual has the latent capacity to commit the act without any social support, and at the intersection of the appropriate forces (aging, alcohol, fatigue, emotional stress) a sexual contact with a child occurs. The model of the behavior that seems to be most useful is that of the releasing mechanism. The pedophile has learned in childhood a certain set of responses that gives him the potential for the act as an adult. Under the appropriate circumstances the potential is realized.

Other sexual acts of the third type that have some of the aspects of pedophilia are homosexual pedophilia, exhibitionism, compulsive peeping, and sex offenses involving incest and the use of force. In the cases of peeping, assaultive offenses, and incest there are certain complicating social or cultural factors. Peeping and the use of minimal aggression in sexual relations are not necessarily viewed as abnormal or deviant. The act of peeping is something that many males have performed when confronted by a lighted unshaded window; however, for the impulse to invade and dominate the sexual life of the individual to such an extent that he must seek out windows in which to look suggests that more than a simple heterosexual interest is involved. A certain amount of aggression by males is expected in the process of sexual exploration in Western culture, and it is clear that excessive force is used more frequently than is reported to the agencies of law enforcement. This is predominantly a lower-class phenomenon, but not totally so, since there is evidence for its occurring among college-level males as well (Kirkpatrick & Kanin 1957). There are indications that those college-level males who do use force have certain other characteristics that differentiate them from males who do not use force. The aggressive male is commonly responding to his own needs rather than to the interactional situation in which he is involved. What may be involved is a blunting of the capacity to recognize negative signs displayed by the female, either as a function of the male’s specific personality characteristics or as a function of his sexual excitement.


The problem of sexual contact between relatives is confined primarily to father-daughter contacts and to a lesser extent to those between brothers and sisters and between cousins. The last two are usually transitory childhood contacts or those that take place just after puberty, and in the case of brother-sister incest they may be made more complicated by the presence of mental deficiency. Father-daughter contacts are most prevalent in rural, or rural-origin, families. The families in which such contacts occur commonly are extremely disorganized families in which the father converts the daughter into a source of sexual and often pseudo-marital relations (Kaufman et al. 1954). Mothers in these families retreat from the center of the family life, and only if the daughter is quite young does the mother inform the police. With very young daughters the offense often has elements of pedophilia.

Sex-offense rates

A useful point to be made about deviant behavior of this third type is that over long periods of time it represents a relatively stable proportion of all offenses known to the police, of all persons arrested, and of all felons in penal institutions. In addition, the rates per hundred thousand in the population have remained stable for many years. The stability of these rates indicates that disturbed and disorganized families represent a rather constant source of individuals who grow up with strategic weaknesses in their character structures and thus become sexual of fenders. The rate of increase in the number of such persons, however, does not exceed the growth rate of the population, indicating that the roots of these kinds of deviant acts are in the family, as the basic unit of social organization, rather than in secondary social structures.

As can be seen from the foregoing discussion, the forms of sexual behavior that are approved arise from sources very similar to those from which the forms called deviant arise, and the margins between one and the other are sharp in moral evaluation but vague in fact. The general situation of sexuality in this society may be best described as that of pluralistic ignorance, a condition of noninformation and misinformation that serves to direct (or misdirect) a powerful drive.

For the vast majority of deviant sexual acts, negative sanctions are frequently expressed but only sporadically enforced. For a smaller proportion of acts, deviant behavior involves recognizable social organization that serves either as a learning structure or as a performance structure for the development and maintenance of the behavior. Finally, for a very small proportion there seem to be no social structural components to the behavior, either for its development or for its continuation, and its origins seem most likely to be in early familial life, as was explained above.

With the general absence of mediating social structures between the original development of sexual capacity in childhood and the performance of sexual roles in adulthood the conditions described above are not surprising. Sexual roles are worked out on an individual basis with a general incapacity to share and reformulate very large portions of the sexual experience. Since there is a high premium on silence in sexual matters, there is rarely an opportunity for social checks on deviant impulses, nor does there exist any system by which the behavior of individuals, if it varies from the rigid normative standards, can be examined in terms of the behavior’s relevant social and psychological meaning. Of all of the areas of deviant behavior, it is sexual deviance which most strongly maintains its “individualistic” character.

John H. Gagnon

[See alsoDeviant behavior; Mental disorders; and the biographies of Freud; Kinsey.]


Achilles, Nancy 1967 The Development of the Homosexual Bar as an Institution. Pages 214-228 in John H. Gagnon and William Simon (editors), Sexual Deviance. New York: Harper.

Bell, Robert R.; and Buerkle, Jack V. 1961 Mother and Daughter Attitudes to Premarital Sexual Behavior. Marriage and Family Living 23:390-392.

Brown, Daniel G. 1958 Inversion and Homosexuality. American Journal of Orthopsychiatry 28:424-429.

Davis, Kingsley 1961 Prostitution. Pages 262-288 in Robert K. Merton and Robert A. Nisbet (editors), Contemporary Social Problems. New York: Harcourt.

Dean, Robert B.; and Richardson, Harold 1964 Analysis of MMPI Profiles of Forty College-educated Overt Male Homosexuals. Journal of Consulting Psychology 28:483-486.

Ehrmann, Winston 1959 Premarital Dating Behavior. New York: Holt.

Ford, Clellan S.; and Beach, Frank A. 1951 Patterns of Sexual Behavior. New York: Harper.

Freud, Sigmund (1905) 1953 Three Essays on the Theory of Sexuality. Volume 7, pages 123-245 in The Standard Edition of the Complete Psychological Works of Sigmund Freud. London: Hogarth; New York: Macmillan. → First published as Drei Abhandlungen zur Sexualtheorie.

Freund, Kurt 1963 A Laboratory Method for Diagnosing Predominance of Homo- or Hetero-erotic Interest in the Male. Behavior Research and Therapy 1:85-93.

Gagnon, John H.; and Simon, William (editors) 1967 Sexual Deviance. New York: Harper.

Giallombardo, Rose 1966 Society of Women: A Study of a Woman’s Prison. New York: Wiley.

Halleck, Seymour L.; and Hersko, Marvin 1962 Homosexual Behavior in a Correctional Institution for Adolescent Girls. American Journal of Orthopsychiatry 32:911-917.

Hampson, John L.; and Hampson, Joan G. 1961 The Ontogenesis of Sexual Behavior in Man. Pages 1401-1432 in William C. Young (editor), Sex and Internal Secretions. 3d ed. 2 vols. Baltimore: Williams & Wilkins.

Harlow, Harry F. 1959 Love in Infant Monkeys. Scientific American 200, no. 6:68-74.

Hooker, Evelyn 1957 The Adjustment of the Male Overt Homosexual. Journal of Projective Techniques 21:18-31.

Hooker, Evelyn 1958 Male Homosexuality in the Rorschach. Journal of Projective Techniques 22:33-54.

Hooker, Evelyn 1966 The Homosexual Community. Pages 354-364 in James O. Palmer and Michael J. Goldstein (editors), Perspectives in Psychopathology. New York: Oxford Univ. Press.

Indiana University, Institute for Sex Research 1965 Sex Offenders: An Analysis of Types, by Paul H. Gebhard, J. H. Gagnon, W. B. Pomeroy, and C. V. Christenson. New York: Harper.

Kallman, Franz J. 1952 Comparative Twin Study on the Genetic Aspects of Male Homosexuality. Journal of Nervous and Mental Disease 115:283-298.

Kallman, F. J. 1963 Genetic Aspects of Sex Determination and Sexual Maturation Potentials in Man. Pages 5-18 in Conference on Community Mental Health Research, 4th, Washington University, St. Louis, 1962, Determinants of Human Sexual Behavior. Edited by George Winokur. Springfield, III.: Thomas.

Karpman, Benjamin 1954 The Sexual Offender and His Offenses. New York: Julian Press.

Kaufman, Irving; Peck, A. L.; and Tagiuri, C. K. 1954 The Family Constellation and Overt Incestuous Relations Between Father and Daughter. American Journal of Orthopsychiatry 24:266-277.

Kinsey, Alfred C. 1941 Criteria for a Hormonal Explanation of the Homosexual. Journal of Clinical Endocrinology 1:424-428.

Kinsey, Alfred C.; Pomeroy, W. B.; and Martin, C. E. 1948 Sexual Behavior in the Human Male. Philadelphia: Saunders.

Kinsey, Alfred C. et al. 1953 Sexual Behavior in the Human Female. Philadelphia: Saunders.

Kirkpatrick, Clifford; and Kanin, Eugene 1957 Male Sex Aggression on a University Campus. American Sociological Review 22:52-58.

Lambertl, Armand 1963 A Type of Peer Relationship in a Girls’ Training School. Menninger Clinic, Bulletin 27:200-204.

Leznoff, Maurice; and Westley, William 1956 The Homosexual Community. Social Problems 3:257-263.

Lindner, Robert M. 1948 Sexual Behavior in Penal Institutions. Pages 201-215 in Albert Deutsch (editor), Sex Habits of American Men: A Symposium on the Kinsey Report. Englewood Cliffs, N.J.: Prentice-Hall.

Marmor, Judd (editor) 1965 Sexual Inversion: The Multiple Roots of Homosexuality. New York: Basic Books.

Money, John 1961 Sex Hormones and Other Variables in Human Eroticism. Pages 1383-1400 in William C. Young (editor), Sex and Internal Secretions. 2 vols. 3d ed. Baltimore: Williams & Wilkins.

Money, John 1963 Factors in the Genesis of Homosexuality. Pages 19-43 in Conference on Community Mental Health Research, 4th, Washington University, St. Louis, 1962, Determinants of Human Sexual Behavior. Edited by George Winokur. Springfield, III.: Thomas.

Ovesey, Lionel 1954 The Homosexual Conflict: An Adaptational Analysis. Psychiatry 17:243-250.

Ovesey, Lionel 1956 Masculine Aspirations in Women: An Adaptational Analysis. Psychiatry 19:341-351.

Reiss, Albert J. JR. 1960 Sex Offenses: The Marginal Status of the Adolescent. Law and Contemporary Problems 25:309-333.

Reiss, Albert J. JR. 1961 The Social Integration of Queers and Peers. Social Problems 9:102-120.

Reiss, Ira L. 1960 Premarital Sexual Standards in America. Glencoe, 111.: Free Press.

Rieff, Philip 1959 Freud: The Mind of the Moralist. New York: Viking.

Sears, Robert R.; Maccoby, E. E.; and Levin, H. 1957 Patterns of Child Rearing. Evanston, 111.: Row, Peterson.

Simon, William; and Gagnon, John H. 1967 The Lesbians: A Preliminary Overview. Pages 247-282 in John H. Gagnon and William Simon (editors), Sexual Deviance. New York: Harper.

Society of Medical Psychoanalysts 1962 Homosexuality: A Psychoanalytic Study, by Irving Bieber et al. New York: Basic Books.

Spitz, Rene A. 1949 Autoeroticism. Psychoanalytic Study of the Child 3/4:85-120.

Sutherland, Edwin H.; and Cressey, Donald R. (1924) 1960 Principles of Criminology. 6th ed. Philadelphia: Lippincott. → First published as a textbook under the title Criminology, with Sutherland as the sole author.

Sykes, Gresham M. 1958 The Society of Captives: A Study of a Maximum Security Prison. Princeton Univ. Press.

Trilling, Lionel (1950) 1957 The Kinsey Report. Pages 223-242 in Lionel Trilling, The Liberal Imagination. Garden City, N.Y.: Doubleday.

Ward, David A.; and Kassebaum, Gene B. 1965 Women’s Prison: Sex and Social Structure. Chicago: Aldine.

Wheeler, Stanton 1960 Sex Offenses: A Sociological Critique. Law and Contemporary Problems 25:258-278.

Whyte, William F. 1943 A Slum Sex Code. American Journal of Sociology 49:24-31.

Young, William C. (editor) 1961 Sex and Internal Secretions. 2 vols. 3d ed. Baltimore: Williams & Wilkins. → Pagination is continuous.

Young, William C.; Goy, Robert W.; and Phoenix, Charles H. 1964 Hormones and Sexual Behavior. Science 143:212-218.


Homosexual practices are among the most ancient manifestations of human sexuality. Abundant evidence for this fact is found in both the religious and the secular literature of the oldest civilizations and in their graphic art as well. Social attitudes in different cultures apparently varied then, as now, from strong condemnation of all homosexual practices to tolerance or permissiveness for some. From the beginning of the Christian era in western Europe ecclesiastical and secular law sought to prevent, control, and eradicate homosexuality by means of severe moral and legal condemnation. By the late nineteenth century, legal penalties were less severe as a result of humanitarian reform, but moral and religious censure had not diminished. That the scientific study of homosexuality began in this period of western European history, and in this cultural climate, is a tribute to the scientific and humanitarian interests of physicians, particularly such pioneers as Richard von Krafft-Ebing, Sigmund Freud, Havelock Ellis, and Magnus Hirschfeld. Their theories were largely concerned with etiology and were based on clinical observations and influenced by developments in the biological sciences, including evolutionary theory. Although by this time cross-cultural observations of homosexuality were also accumulating, together with the rudiments of a general theory, they had little effect on the mainstream of research, which continued to focus on the etiology of homosexuality among individuals of Western culture.

Increased research on homosexuality in Western countries since World War II has been coupled with more open discussion of the subject. Although there is no reliable evidence that the incidence of homosexuality is increasing, some forms of it are more socially visible. In part, this may be accounted for by the publicity given to scientific investigations of homosexuality and to discussions of the controversial social, legal, moral, and mental health issues related to it. Emergent homosexual “communities” in large urban centers and “homophile” organizations openly protesting public policy further contribute to this heightened visibility and increased public awareness. These developments reflect radical changes in the attitudes of the public and of homosexuals, in the social character of homosexuality, and in the volume and focus of scientific research.

Homosexuality (homo- derives from the Greek root meaning “same") includes an extraordinary diversity of dyadic relations and of individual mental states and action patterns. The patterns of social organization that develop when homosexuals seek each other out also vary greatly. Because of this diversity, “homosexuality” is an ambiguous term with many meanings. Some investigators, such as Ford and Beach (1951), limit the term to overt sexual relations between individuals of the same sex. For others, notably Kinsey and his associates (1948; 1953), the degree of psychic arousal and the frequency of overt sexual response to individuals of the same or opposite sex determine ratings on a heterosexual-homosexual continuum that ranges from exclusively heterosexual to exclusively homosexual, with several intermediate ratings. Investigators with a clinical perspective (see, for instance, Marmor 1965, p. 4) frequently stress motivational and subjective aspects of erotic preference as criteria for defining persons as homosexual; overt behavior is considered as of secondary importance. In the social perspective of homosexual subcultures, the defining criteria may be shared understanding of sexual preferences for members of one’s own sex and participation in social activity centered on the search for, and interaction with, these individuals (Goffman 1963, pp. 143-144). Thus, it is clear that “Who is homosexual?” and “What is homosexuality?” are very complex questions, clarification of which would be a lasting contribution to social science.

Research on homosexuality is also of major significance for the light it throws on the relation of social structure and cultural norms to patterns of human sexuality; on the origins, development, and essential features of normal and abnormal personality; on masculinity and femininity; and on social deviance and deviant subcultures.

Etiology and determinants

In those Western societies that define homosexuality as deviant, there are four major theoretical issues concerning the etiology and determinants of persistent or predominant adult homosexuality. (1) Is the human organism psychosexually neutral at birth, so that learning processes determine homosexual object choice in adults, or are there “inherent sexual predispositions” (Diamond 1965, p. 168) which selectively influence the effects of learning? (2) What is the nature and content of the learning processes by which homosexual object choice develops? Is the appropriate developmental model a deviant role or a personality system with intrapersonal traits, motives, and gender identifications incompatible with the social-sexual capabilities and self–other expectancies of adult relations with the opposite sex? Does positive conditioning of sexual responses to persons of the same sex, or negative conditioning to persons of the opposite sex, or a combination of both, account for homosexuality? (3) Are particular periods in the developmental process, such as early childhood or adolescence, critical for homosexual object choice? (4) Are parent-child relationships in the nuclear family crucial in determining whether an individual becomes homosexual, or are peer relationships in childhood and adolescence, and deviant subcultures in adolescence and early adult life, of equal or possibly greater importance? These four issues are highly controversial and cannot be resolved by the research evidence currently available. A brief historical review of some of the major studies will illustrate differences in theoretical assumptions, in methodological approaches, in findings, and in conclusions, and it will indicate important areas for continued research.

Theories of homosexuality

Early theorists stressed the importance of biological predispositions as determinants. Krafft-Ebing (1886) theorized that homosexuality was produced by a dominance of opposite-sex brain centers (female centers in males or vice versa) or that it was a hereditary disease. Freud (1905) postulated constitutional bisexuality as one of the biological predispositions which, together with early learning experiences or fixations, determined homosexual object choice. The major studies of biological predispositions have focused on genetic, chromosomal, and hormonal anomalies or dysfunctions. Comparisons of predominantly homosexual and heterosexual males with respect to chromosomal abnormalities have resulted in negative findings. In several studies, maternal age and birth order of homosexuals in a sibling series were significantly later than in control groups of heterosexual men. The degree of concordance of homosexuality in identical male twins appears to be significantly higher than in fraternal twins (for a review of these studies, see Pare 1965). The evidence is controversial but suggests that genetic mechanisms may be important. Significant progress could be made by replicating the studies of twins, birth order, and maternal age and by an intensified search for genetic determinants. Homosexual object choice in adults is not affected by hormone therapy and differences between the endocrine systems of homosexuals and those of heterosexuals have not been reliably demonstrated (Perloff 1965). As more accurate techniques of measuring hormonal activity develop, continued research may be valuable. A potential contribution of biological factors to human homosexuality is suggested by recent studies of animals, in which the location and functions of sex-specific behavior centers in the brain have been demonstrated (Krafft-Ebing may be vindicated!). Studies at the infrahuman level have also demonstrated that hormones injected prenatally produce effects on adult sexual behavior (for a review of these studies, see Diamond 1965). Thus, specific biological predispositions may be crucial variables, interacting with psychological and cultural variables, especially in producing the highly feminized males and masculinized females often referred to as “inverts.”

The experiential determinants emphasized by classical psychoanalytic theory are pathogenic parent-child relations and failure to successfully resolve (unconsciously) the vicissitudes of the three basic phases of libidinal and object-relations development in the critical period of infancy and early childhood. Some revisions of psychoanalytic theory reject constitutional or psychological bisexuality and emphasize ego adaptations to environmental difficulties. With few exceptions, however (see, for instance, Thompson 1947; Sullivan 1953), pathogenic relations with parents in early childhood are assumed to be the crucial determinants, although individual patterns of behavior are subject to modification during adolescence. (For a review of psychoanalytic theories, see Society of Medical Psychoanalysts 1962, chapter 1.)

This theoretical assumption currently relies for empirical support on conflicting and controversial data derived largely from comparing the life histories of adult homosexual and heterosexual males. Freund and Pinkava (1961) carefully tested clearly formulated hypotheses concerning parent-child relations and homosexuality by comparing interview data from male homosexual patients in a psychiatric hospital with similar data from two contrasting matched groups: neurotic heterosexual patients in the same hospital and nonpsychiatric patients. They found no significant differences between these groups with respect to the dominance, control, or decisive influence of either parent in the family. Relations between homosexual sons and their fathers were more “disturbed,” a finding tentatively interpreted by the authors as secondary to, rather than a cause of, homosexuality, since the disturbance appeared to be mainly on the son’s side of the relationship. Reports of greater conflict between parents of homosexuals signified a lack of tolerance by their sons for marital conflict rather than greater incidence of actual conflict. Bieber and his co-workers (see Society of Medical Psychoanalysts 1962) compared questionnaire data provided by psychoanalysts concerning their homosexual and heterosexual patients. Some differences between patient-mother-father, mother-son, and father-son relationships in the contrasting groups were statistically significant. The authors concluded that the chances of becoming homosexual or of developing severe homosexual problems appear to be high for any son who is exposed to the combination of a “close-binding-intimate” mother who dominates and minimizes her husband, and a father who is detached or hostile (p. 172). Schofield (1965) compared interview data obtained from six matched groups: (1) homosexual and pedo-philiac prisoners convicted of homosexual offenses, (2) homosexual and heterosexual patients who were in psychiatric treatment, (3) homosexuals and heterosexuals who were neither in treatment nor in prison. The proportion of individuals from disturbed home backgrounds was similar in each matched pair of groups; it was lowest in the non-treatment-prison groups and highest in the patient groups. The proportion of individuals who reported possessive, overprotective, or dominant mothers and poor or nonexistent relations with fathers in childhood was larger in the combined homosexual groups (pedophiliacs excluded) than in the combined heterosexual groups (p. 105).

The evidence from these and many similar studies does not support the assumption that pathological parent-child relations are either necessary or sufficient antecedents or determinants of adult homosexuality. The evidence does indicate, however, that some forms of familial pathology appear to be associated with increased vulnerability of some individuals to homosexual development, and it suggests that psychopathology is more frequently associated with homosexuality in these individuals. Longitudinal studies could now be designed for the testing of hypotheses specific to the differential effects of variable combinations of parent-child relations on the development of adult psychosexual patterns. The potential contribution of such studies is very high.

Empirical evidence that determinants in adolescence are critical for adult homosexuality is fragmentary. The theoretical rationale for this assumption is that one of the major developmental tasks of adolescence is the achievement of a stable psychosexual identity. An extensive body of clinical-theoretical literature emphasizes the adolescent period as critical for the resolution of intrapsychic psychosexual conflicts originating in infancy or early childhood (Society of Medical Psychoanalysts 1962). Sullivan (1953) stressed very close and intense preadolescent friendships with peers of the same sex as one of the essential conditions for heterosexual development and the separation of lust and intimacy “dynamisms” as one precondition of adult homosexuality.

Assuming that the human capacity for homosexual as well as heterosexual response is a basic part of the mammalian heritage, Kinsey and his associates (1948) suggested that cultural pressures and social conditioning may determine the final object choice. That adolescence may be a critical period for males is indicated by (1) the existence of a high positive relation between the early onset of adolescence and frequency of homosexual activity during adolescence and later life (ibid., p. 630) and (2) occurrence of the peak of sexual activity between ages 16 and 20. Among females, however, no consistent relation was found between homosexual activity and early onset of adolescence (Kinsey et al. 1953, p. 462), and the peak of female sexual activity occurs in the late twenties (ibid., p. 759).

Group norms and peer relationships in adolescence may determine the relative frequencies of heterosexual and homosexual patterns in adults by affecting their self concepts, sex role expectations and performances, and notions of what constitute the boundaries of permissible behavior. Reiss (1961) found that members of a group of male juvenile delinquents who engaged in sexual acts with adult homosexuals did not define themselves as homosexuals and did not continue homosexual activity as adults. Group norms defined the limits of permissible sexual activity, rules of payment, frequency of contact, and derogatory attitudes toward homosexuality. Conversely, late adolescents or young adults who participate in homosexual cliques or informal homosexual groups may progressively redefine homosexual behavior in positive terms, defining themselves as homosexuals in the process, and thus beginning a homosexual “career” (Hooker 1965a, pp. 90, 101).

Adult homosexual roles may be formed by a continuous process of social-sexual learning, from early childhood to adolescent and early adult life. Self-other interactions, including overt sexual activities, in deviant socially marginal pairs or groups of peers of the same sex, may be critical determinants of motives, self concepts, sex role expectations and abilities, and performances that are congruent with deviant and homosexual roles and incongruent with heterosexual roles. At critical choice points in the developmental process or continuously from early childhood, situational and social-structural variables may produce effects influencing adult patterns—for instance, the feminizing effects of older sisters on younger brothers, situationally determined isolation from peers of the same sex in childhood and adolescence, or isolation from peers of the opposite sex in adolescence.

From the limited evidence currently available, it is clear that the diverse forms of adult homosexuality are produced by many combinations of variables, including biological, cultural, psychodynamic, structural, and situational. No single class of determinants, whether psychodynamic, cultural, or biological, accounts for all or even one of these diverse forms. The relative importance of each kind of determinant apears to vary greatly from one individual to another. It should be noted, however, that the classification of types of homosexuality with respect to patterns of determinants and characteristic adult features is an important problem for research. It is also clear that many developmental models are required to account for the diversity of developmental sequences leading to confirmed patterns of adult homosexuality. The continued search for determinants may be moreproductive if pursued in the perspective of development over time, rather than in the traditional perspective of origins or causes.

Psychodynamics of adult homosexuality

Psychoanalytic theories assume that homosexuality is a symbolic expression of unconscious psychodynamic mechanisms and that in any given case these mechanisms and symbolizations are complex and numerous. Thus, according to Freud 1905), male homosexuality may be a narcissistic search for a love object symbolizing the self; an avoidance and derogation of females induced by unconscious castration anxiety and fear of them, coupled with primary narcissism focused on the male genitals; and an unconsciously incestuous and therefore forbidden attraction to women associated with unconscious castration anxiety and fear of punishment by males. Stronger identification with the mother is symbolically expressed in “feminine” sexual submission to males; that with the father, in “masculine” domination of males. Freud postulated similar complex symbolic expressions of unconscious psychodynamics for female homosexuality; the primary difference between male and female homosexuality, he thought, related to the nature of castration anxiety. However, there is a paucity of theoretical assumptions that differentiate male and female homosexuality with respect to personality dynamics and its correlates (see Romm 1965). Ambivalence, chronic anxiety, intense hostility toward the partner, and longing for love characterize female homosexual relationships, according to Wilbur (1965). Thompson (1947) considered homosexuality in both sexes as a symptom of diverse character problems covertly expressed in homosexual relations (among these problems were fear of the opposite sex, need to defy authority, fear of adult responsibility, and destructiveness of self or others). These theories involve complex patterns of variables that are difficult to state as clear hypotheses. Consequently, the design of an appropriate methodology for testing them presents problems. Research studies have therefore focused on selected aspects of the theories or variables for which clinical impression or theoretical rationale provides justification.

Masculinity and femininity

A major focus of research studies on male homosexuality is the complex issue of psychological masculinity-femininity and its relation to gender identification. Terman and Miles (1936), using masculinity-femininity test scores and interview data, compared “passive” male homosexuals (defined by “female” roles in sexual acts) with “actives” (defined by “male” roles) and then compared both of these with (presumably) selected groups of heterosexual men and women. The femininity scores of the passives were higher than those of any other male group tested. The masculinity scores of the actives were higher than those of the passives and of a group of regular-army men. These results suggested that individuals in both homosexual groups represented deviations in masculine and feminine identifications, interests, activities, and feelings. Interview data on the passives supported the masculinity-femininity test results. Similar results were obtained on a small sample of active female homosexuals.

In subsequent studies of masculinity-femininity, data on sexual practices and preferences are rarely reported, but the evidence from research on other dimensions of male homosexuality indicates that samples of homosexuals usually include many individuals whose preferences and practices are highly variable, as well as those representing the two extremes of active and passive (Society of Medical Psychoanalysts 1962; Hooker 1965fc; Schofield 1960). The distribution of the scores of homosexuals and control groups of heterosexuals on the masculinity-femininity scale of the Minnesota Multiphasic Personality Inventory (MMPI) indicates that homosexuals are significantly more “feminine” in their responses than heterosexuals. Thus, Doidge and Holtzman (1960, p. 11) found that a group of homosexual men in the Air Force who had been investigated for homosexual offenses scored significantly higher (that is, were more “feminine") than three control groups of predominantly or exclusively heterosexual men. Braaten and Darling (1965, p. 298) found that, while overt homosexual males (including exclusives and bisexuals) and covert homosexuals who were patients in a college clinic did not differ significantly on masculinity-femininity scores, as a total group (overts plus coverts) they scored significantly higher than a control group of equally disturbed, nonhomosexual patients. Although a group of homosexual males who were not in treatment and were apparently functioning effectively in the community scored significantly higher than a control group of heterosexual males on the masculinity-femininity scale of the MMPI and the femininity scale of the California Personality Inventory (CPI), significant differences were not found in scores on various projective measures designed to assess unconscious gender identification (unpublished research by the author). Feminine identification and confused or unstable gender identification in male homosexuals have also been assessed by the use of content indices in the Rorschach Test and the Thematic Apperception Test (TAT). The interstudy reliability of these indices in differentiating homosexual and heterosexual groups is low (Hooker 1965a; Lindzey 1965). Chang and Block (1960), using a list of adjectives, derived identification scores from descriptions by male homosexual and heterosexual research subjects of the ideal self, the mother, the father, and their actual selves. The results supported the hypothesis that homosexual males were more strongly identified with the mother and did not identify with the father.

Conclusions from these findings must be drawn with caution. Psychological masculinity and femininity are not unitary dimensions of personality but composites of different clusters of interests, attitudes, and feelings. The test measures are also multivariate in character. Thus, high “femininity” scores may indicate ego sensitivity and denial of interest in culturally masculine occupations (Braaten & Darling 1965). Research findings currently available do not provide even roughly sketched outlines of the complex patterns of psychological masculinity-femininity and gender identification in male or female homosexuals. If current measures of these variables are to remain in use, the multivariate character of what is being measured must be taken into account. Also, new and more appropriate methods must be devised for assessing different clusters or profiles of the conscious, unconscious, cultural, and biological components of masculinity, femininity, and gender identification. The selection of samples of research subjects must make allowance for the broad range and diversity of masculine and feminine patterns and the variable combinations of these.

Other personality dimensions

The evidence supporting assumptions related to other psycho-dynamics or personality correlates is equally inconclusive. Bieber and his co-workers (Society of Medical Psychoanalysts 1962) concluded that their findings provide convincing support for the assumption that homosexuality is a specific form of sexual psychopathology: fear of the opposite sex or of heterosexual expression. But none of the evidence, such as fear of injury to the genitals, fear and aversion to female genitalia, or anxiety associated with heterosexual behavior, was specific to the homosexual group. The evidence cited for the conclusion that other psychopathologic processes in personality are secondary but invariably occur in homosexuals is subject to the same limitations. Moreover, the research subjects were patients in psychoanalytic treatment, so that the relation between psychopathology and homosexuality was confounded. Braaten and Darling (1965) found no evidence to support assumptions that homosexuals, as rated by psychiatric interviewers, were more psychopathic, paranoid, or schizoid than a control group of heterosexuals and found no differences in psychopathology as measured by the MMPI. Psychiatric ratings on narcissism and masochism were higher for homosexuals but not specific to them. The significantly higher scores of homosexuals on all clinical scales of the MMPI were interpreted by Doidge and Holtzman (1960) as evidence that homosexuality is a symptom of severe personality disturbance. In a number of studies of male homosexuals who were not in treatment or in disciplinary settings, the results obtained by the use of the MMPI, TAT, Rorschach, and other psychological measures did not justify the conclusion that homosexuality is necessarily and invariably a concomitant or symptom of psychopathology. In many individuals no evidence of psychopathology was found (Hooker 1957; 1965a). Nevertheless, the assumption that a modal pattern of personality structure and psychodynamics characterizes male homosexuals is explicitly stated or is implicit in the methodology used in many research studies. Having been unable to establish any differences between a group of homosexual males and a control group of heterosexuals, either in degree of self-acceptance or in ideal-self descriptions, Chang and Block commented that since “homosexuals and normals are quite obviously rather different sorts of people,” their data must be regarded as questionable (1960, p. 309). Nevertheless, the only obvious difference between homosexuals and heterosexuals is in psychosexual object choice. All experienced clinicians and research workers report that the personality differences among individual homosexuals are far more apparent than the similarities. Investigators who include the data on individual differences in their studies have found a great diversity of personality patterns. A singular exception is the finding of Cattell and Morony (1962) that low ego strength, high extroversion, low superego, and a radical social outlook characterize the personality profiles of male homosexuals. They conclude that it is legitimate to speak of the homosexual-personality profile.

The inconclusiveness of research on homosexuality is a function of many variables, including differences in investigators, in samples of research subjects and situational settings, in methodology, and in concepts of homosexuality. Lindzey (1965) has demonstrated the significant influence of many of these variables in a study designed to cross-validate objective indices of homosexuality in the TAT and to test the predictive efficiency of clinical methods of diagnosing homosexuality compared with actuarial ones. The limitations of psychological tests in the diagnosis of homosexuality have been documented in many studies. Inferences from these measures and generalizations about the personality characteristics of homosexuals must be made with caution and within the limits set by the characteristics of the research sample, the setting, and the methodology.

With few exceptions, studies of the personalities of homosexuals have been stated in terms of dichotomous attributes—psychopathology versus normality, specific sexual psychopathology versus pervasive personality disorder, feminine versus confused identification, feminine versus masculine identification, and other pairs of mutually exclusive variables. Thus, in effect, homosexuality has been treated as a clinical entity. But contemporary personality theories are richly suggestive of dimensions or subsystems of personality that are potentially relevant to psychosexual patterns, and until now, these theories have scarcely been tapped as sources of hypotheses. Some of the effects of a changing climate of public attitudes toward homosexuality and of an increasing freedom to conduct scientific studies of human sexuality in all of its complex psychological, social, and biological aspects are manifested in the small but growing number of behavioral scientists whose interests and competence in personality research are focused on homosexuality. With the cooperation of homosexuals in the “homophile” and larger communities, it is no longer necessary to depend on subjects in clinic or prison settings. With the use of more rigorous criteria in selecting homogeneous samples of subjects and of more appropriate research designs, including cluster and profile analyses, it may be possible to differentiate personality patterns among homosexuals. Also, progress may be achieved in resolving the long-debated issue as to whether some homosexuals differ from their heterosexual counterparts only in sexual object choice or are invariably different in some personality dimensions.

Role differentiation and typology

The perspectives of the two-sexed heterosexual society have so dominated the study of homosexuality that typologies of homosexual and related roles have been developed only to a very limited extent. Thus, Ferenczi (1914) distinguished between male inverts who have pervasive feminine identifications and assume a passive sexual role and males who assume an active sexual role and are apparently masculine in every respect except sexual object choice. Many theoreticians and empirical investigators (see, for instance, Terman & Miles 1936), while emphasizing the active-passive dichotomy in sex roles and the corresponding masculine-feminine dichotomy in gender identifications, have noted that the patterns of many male homosexuals represent variable combinations and intermediate grades. Recent studies (Schofield 1960; Society of Medical Psychoanalysts 1962; Hooker 1965b) have shown that the terms “activity” and “passivity” may be useful in referring to psychological states but are inaccurate and misleading when applied to sex roles and should therefore be discarded. The terms “insertee” and “insertor” (Society of Medical Psychoanalysts 1962) are more accurate and operationally useful equivalents. Studies of the sex-role preferences and practices of male homosexuals indicate that a large proportion of them habitually perform either role, according to their own or their partners’ momentary preference. For the majority, consciousness of male identity and of masculinity or femininity appears to show no clear relation to sex roles. Sustained dyadic relations and the shared perspective of male homosexual subcultures appear to increase the flexibility of social–sexual roles and to create the social conditions for the emergence of new solutions to the problems of role differentiation in a one-sex world (Hooker 1965b).

Studies of male homosexuality in correctional institutions (Huffman 1960; Sykes 1958) suggest that homosexual social-sexual roles are clearly defined by inmate codes. According to these codes, the masculinity of the individual who assumes the insertor role in a homosexual act is not diminished and may even be increased. Within the institution, the role carries no implication that the individual is homosexual. Emotional involvement with a partner appears to be rare. Gratification of needs for power, status, and sexual release is presumably legitimate motivation for assuming the role. Individuals who assume insertee roles occupy social positions of inferior status to which stereotyped attributes of femininity and homosexuality are ascribed. These findings apply only to institutions in which known homosexuals are not segregated from the rest of the prison population.

As noted above, the literature of female homosexuality is fragmentary in comparison with that of male homosexuality. Studies of role differentiation have been conducted largely in correctional settings for adolescent girls or prisons for adult women. In prisons for women a male-female dichotomy in social–sexual roles develops and is sustained with little flexibility or variation (Ward & Kassebaum 1965). The masculine, or “butch,” role is characterized by emulation of the male in appearance, mannerism, and aggressiveness in social interaction as well as in sexual relations. The feminine role is identical in the prison and in the larger society, except for the sexual partner. The woman who assumes the masculine role becomes emotionally involved with her feminine role partner and defines herself as homosexual more frequently, and receives sexual gratification less frequently, than her counterpart in the male prison. Whether the clear-cut male-female role differentiation in prison settings is also characteristic of female homosexual patterns in the larger community is unknown. Despite the differences, the similarities of homosexual role differentiation in men’s and women’s prisons are striking. It appears that individuals in both settings who assume opposite-sex roles are more likely to have been overt homosexuals before entering prison and to continue homosexuality on being released. In studies of correctional settings for adolescent girls, it has been found that the roles assumed in the course of homosexual relationships represent variations of all family roles.

The empirical data obtained in the restricted environments of correctional institutions for men and women support a dichotomous typology of homosexual roles in which the male and female social–sexual roles in the larger community are reflected. However, male homosexual role patterns in other cultural settings appear to show extensive variations for which a taxonomy, or system of classification, is urgently needed. Studies of female homosexual role patterns in nonprison settings (several were in progress in 1966) can provide the necessary data for a comparison of variability within and between the sexes. The development of taxonomies of male and female homosexual patterns is a theoretical problem of paramount importance. To be meaningful, such taxonomies must include variations in many dimensions of personality, in sexual and social behavior, and in the social settings or subcultures in which many homosexuals participate.

Demographic and cultural variation

In undertaking empirical research on the incidence or demographic distribution of homosexuality in urban industrialized societies, investigators are confronted with two major questions: (1) Who or what is to be counted as homosexual? and (2) How can representative samples be obtained in different social strata of individuals whose behavior is socially, morally, and legally stigmatized? In view of the magnitude of these problems, it is not surprising that few have even attempted to solve them. The most systematic research studies of the incidence and social correlates of homosexuality are those of Kinsey and his associates (1948; 1953), who rejected the counting of homosexuals as an inappropriate and impossible task. Accumulative estimates of incidence were based on the number of individuals brought to orgasm by physical contact with others of the same sex during specified periods. They attempted to collect interview data from samples of equal size, varying in demographic characteristics. The samples (5,300 white males, 5,940 white females) were neither representative of the total population nor randomly selected from it.

For white males the estimated incidence was indicated by the following: (1) 37 per cent had at least one overt homosexual experience between the onset of adolescence and old age; (2) 10 per cent were predominantly homosexual for at least three years between the ages of 16 and 55; (3) 4 per cent were solely homosexual after the onset of adolescence. For white females, the estimated incidence was much lower: (1) 13 per cent had at least one overt homosexual experience between adolescence and old age; (2) 28 per cent had covert or overt experience; (3) a half to a third as many females were primarily or exclusively homosexual in any age period.

The major findings on the social correlates of homosexual activity for males were that the highest incidence occurred among unskilled and semiskilled workers and the lower white-collar class; among those with high school education and of urban background; and among the less devout or inactive church members. The” lowest incidence occurred among the professional and upper white-collar classes and in the college-educated group. However, the incidence among skilled workers was closer to that among the professional and upper white-collar classes than to that among the lower occupational groups. Those with grade school education displayed a level of homosexual activity intermediate between that of high school- and that of college-educated groups. Homosexual activity in females increased with education, so that the incidence was highest among those with college or postgraduate education; for males, the opposite was true. For both males and females the highest incidence occurred among the less devout or least active in church affiliation.

Although the assumptions, methods, and conclusions of these studies are highly controversial, their findings are most frequently cited as evidence for a high incidence of homosexuality in the United States. They are also believed (on the basis of rather limited demographic evidence) to support the assumption that cultural differences between social groups determine the unequal distribution of homosexual activity in different segments of society.

This assumption also appears to be supported by Schofield’s study in England (1960). A group of 127 men who identified themselves as homosexuals to the investigator and were also rated by him as predominantly or exclusively so were interviewed concerning their backgrounds and current social and sexual patterns. The largest proportion of men were in lower occupational levels and the smallest in the professional group, a distribution of men among occupational levels that is similar to the total population’s. However, the number of men in nonmanual and artistic occupations and in the professions exceeded chance expectations, and those employed in a technical capacity and manual workers were underrepresented. In educational level attained, the sample was similar to the total population in that the largest number had the statutory minimum of education. However, the number of those with extensive education and high social status was higher than would be expected in a random sample of the total male population. The majority of these men lived in London, to which a large proportion of them had migrated from other urban centers. Active church affiliation was only slightly lower than that of the general population. In a continuation of this study, fifty men were selected who had been neither convicted of homosexual offenses nor involved in therapy of any kind (Schofield 1965). They were compared with an equal number of men in three groups: homosexuals in prison, homosexuals in therapy, and pedophiliacs convicted of offenses against males. On Kinsey ratings, the prison group was predominantly or exclusively homosexual; the patient group included a small number who were equally homosexual and heterosexual or predominantly heterosexual; a quarter of the convicted homosexuals and 60 per cent of the pedophiliacs were either equally homosexual and heterosexual or predominantly heterosexual. In educational achievement, occupational status, earned income, social class, and upward social mobility, the men who were neither in prison nor in therapy ranked highest and the prison groups lowest. The four groups were differentiated not only in their demographic characteristics but also in their patterns of social and sexual relationships. Thus, the men who had not sought treatment or been arrested were better integrated into the larger community both at work and in their leisure time. They also had larger circles of homosexual friends and more long-standing relationships with other men. Only in this group were there substantial numbers of men who maintained a household with another homosexual. They were far less promiscuous than the patient or prison homosexual groups, although the frequency of sexual experience was equally high.

It is apparent that comparison of the findings in this British study with those in the Kinsey report has limited meaning. However, some similarities and differences are of interest. In both studies, homosexuals were found in all social classes, educational levels, and occupations. A larger proportion of homosexuals (or of homosexual activity) in all studies took place in urban settings and in lower occupational levels and social classes. In educational achievement, however, the proportion was greater in the middle level (high school) in America and in the lower level (secondary modern school) in England. That in both countries the distributions were skewed toward the lower end of the occupational and social class range may well be accounted for by the fact that individuals in higher levels would not volunteer to be interviewed. The same limitation is relevant to the Kinsey finding that the incidence of homosexual activity is lowest among the most devout religious groups. Moreover, in the English studies the incidence of active religious affiliation did not appear to be significantly lower among homosexual males.

In short, whether homosexuals or homosexual activities among males or females are more heavily concentrated in different social strata remains an open question. Whether the social patterns or incidence of those who define themselves as homosexual are different in varied social levels is of equal or possibly even greater importance.

Ethnographic accounts

In a survey of studies of sexual behavior in 190 societies, Ford and Beach (1951) found that ethnographic accounts of homosexual activities were available for 76. In 49 of these some forms were condoned, regarded as normal or socially acceptable, and even encouraged for at least some classes of individuals. In some cultures, such as those of the Siberian Chukchee and the Aleutian Koniaga, an institutionalized role of berdache, or shaman, is provided for adult male homosexuals. These men adopt feminine dress, activities, mannerisms, become “wives” of other men, and assume the “female” role in anal inter-course. Their social status may be high. Male homosexuality in some societies, for instance among the Keraki and Kiwai of New Guinea, is an institutionalized feature of puberty rites. Thus, all males must engage in homosexual practices, either as initiates in the insertee role in anal intercourse or as married or unmarried males who perform the inserter role. Such practices are believed to be essential for male growth and strength. It has also been reported that all men and boys among the Siwans of northeast Africa engage in homosexual practices; married and unmarried males have homosexual and heterosexual liaisons. In 28 of these societies, adult homosexuality was reported to be rare, absent, or only practiced in secret. In all societies (except that of the Sirione in Bolivia) in which homosexuality was said to be rare, specific social pressures were directed against it. Condemnation and penalties ranged from ridicule to threat of death. The unreliability of the estimates of incidence in these societies was apparent. Information concerning female homosexuality was reported for only 17 societies. It appeared highly probable that females were less likely than males to engage in homosexual practices.

In view of the small number of societies for which data are currently available and the questionable reliability and meaning of the data in the many studies that report sexual behavior while ignoring its context of culture patterns and social structures, it is apparent that descriptive or explanatory generalizations concerning demographic and cultural variations in homosexuality are precariously based. Whether a majority of societies consider some form of homosexuality normal and socially acceptable is an important issue for continued cross-cultural research. Clearly, some societies not only consider adolescent or childhood homosexuality normal (some evidence suggests an increasing trend in this direction in the United States) but also regard homosexual relations between adult and adolescent males as important, normal, and socially approved alternatives to heterosexual intercourse (Davenport 1965).

Subcultures and social organizations

Homosexual communities and subcultures in modern urban centers are collective reactions to legal pressures and social stigma. Whether they are largely created by societal reactions to this form of sexual expression (Becker 1963) or, on the contrary, would develop under any sociolegal conditions is debatable (Schur 1965, p. 86). A central thesis in contemporary theories (Lemert 1951; Becker 1963; Goffman 1963) is that deviant as well as conforming behavior is learned in interaction with others and that it cannot be understood without reference to the societal reactions it invokes. Male homosexual collectivities develop on a sizable scale only in modern industrialized societies, which define this form of sexual expression as criminal, pathological, or immoral. In adjusting to this stigmatized status, a large number of homosexuals in social interaction with others who share similar inclinations make it the basis of their social identity and way of life. A larger majority cope with the problems of homosexuality in other ways, whether or not they enter some sector of a homosexual community.

Homosexual communities in large cities are made up of constantly changing aggregates of persons who are loosely linked by friendship and sexual interests in an extended and overlapping series of networks. Some network clusters form tightly knit cliques of friends and homosexually “married” pairs, while in others, informal groups or social organizations develop. The structure of the communication, friendship, and sexual network among members of the community is complex. Community gathering places are centers from which information is transmitted concerning social occasions for homosexuals and the attitudes and organized activities of law-enforcement agencies concerning homosexuals. In some sectors of these networks, bars, taverns, or private clubs are informally institutionalized as homosexual territory, although under surveillance of the police, who sometimes exact tribute for their connivance. These settings provide opportunities for the initiation of sexual contact and for an in-group social life. Some are primarily market places for making sexual contacts, while others are predominantly centers of social activity, although arrangements for sexual activity may occur. A standardized and essential feature of interaction in bars, baths, streets, and parks is the expectation that sex can be had without obligation or commitment. Sexuality is separated from affectional and social life and is characterized by promiscuity, instrumentality, and anonymity. In other sectors of the homosexual community, sexuality is integrated in the affectional, personal, and social patterns of individuals who establish relatively stable and long-lasting relationships.

Many subcultures develop in large homosexual communities, with a distinctive vernacular, distinctive modes of speech and dress, distinctive values and norms regulating sexual and other forms of conduct, and distinctive patterns of interaction with homosexuals and nonhomosexuals. Some individuals may be totally immersed in a homosexual subculture of the community. A wide variety of occupations is created by the demand for services to homosexuals. Male prostitutes, many of whom are homosexuals, constitute one occupation (Reiss 1961; Schur 1965). Entrepreneurs who are homosexual may develop an occupation in arranging social entertainment and large social gatherings for members of a particular homosexual group. Thus, at work and in leisure time, in personal and social life, contact with the larger society may be infrequent. For the majority, however, behavioral, temporal, and spatial separation of the homosexual and heterosexual worlds is required to manage the strategies of passing as heterosexuals at work and in social contact with heterosexuals. In fact, there are many different methods, some of them quite complex, both of participating in the homosexual community without leaving the larger society and of ignoring the larger society altogether. Related to these patterns of integration and separation are the different social levels of homosexual groups.

Female homosexual, or lesbian, communities apparently develop on a smaller scale, with informal groups, cliques, and special gathering places. But a formal organization of lesbians and an official publication with national circulation indicate that the collective aspects of female homosexuality have some importance. No empirical studies of these aspects are currently available. Although homophile organizations, whether male or female, constitute and represent a very small minority of the total homosexual population, they achieve social significance by the role they assume in openly protesting the status assigned to homosexuals by the larger society.

Empirical studies of, and theoretical inquiry into, homosexual subcultures are very limited (Leznoff & Westley 1956; Hooker 1965; 1965b). They suggest that commitment to a homosexual identity and career and the stability of that commitment and the career pattern that follows may be determined by entry into the homosexual community and by the ensuing identity transformation and socialization processes. However, the relation between personality and homosexual subculture variables in determining the commitment to, and patterns of, adult homosexuality is complex. For many the stability of the commitment appears to be a function of the interaction of both sets of variables. Probably a majority of those who are committed to an exclusive homosexual pattern have defined themselves as homosexual before they enter the community. In major urban centers, however, it appears unlikely that adult homosexuals are ignorant of, or indifferent to, the community.

Evelyn Hooker

[See alsoPersonality; and the biographies ofEllis; Freud.]


Reviews of research studies may be found in Marmor 1965; Ruitenbeek 1963; Schofield 1960; 1965. Kinsey et al. 1948; 1953 include surveys of earlier work on homosexuality.

Becker, Howard S. 1963 Outsiders: Studies in the Sociology of Deviance. New York: Free Press.

Braaten, Leif J.; and Darling, C. Douglas 1965 Overt and Covert Homosexual Problems Among Male College Students. Genetic Psychology Monographs 71: 269-310.

Cattell, Raymond B.; and Morony, John H. 1962 The Use of the 16 PF in Distinguishing Homosexuals, Normals, and General Criminals. Journal of Consulting Psychology 26:531-540.

Chang, Judy; and Block, Jack 1960 A Study of Identification in Male Homosexuals. Journal of Consulting Psychology 24:307-310.

Davenport, William 1965 Sexual Patterns and Their Regulation in a Society of the Southwest Pacific. Pages 164-207 in Conference on Sex and Behavior, Berkeley, California, Sex and Behavior. Edited by Frank A. Beach. New York: Wiley.

Diamond, Milton 1965 A Critical Evaluation of the Ontogeny of Human Sexual Behavior. Quarterly Review of Biology 40:147-175.

Doidge, William T.; and Holtzman, Wayne H. 1960 Implications of Homosexuality Among Air Force Trainees. Journal of Consulting Psychology 24:9–13.

Ferenczi, Sandor (1914) 1963 The Nosology of Male Homosexuality. Pages 3-16 in Hendrik M. Ruitenbeek (editor), The Problem of Homosexuality in Modern Society. New York: Dutton. → First published in German.

Ford, Clellan S.; and Beach, Frank A. 1951 Patterns of Sexual Behavior. New York: Harper.

Freud, Sigmund (1905) 1953 Three Essays on the Theory of Sexuality. Volume 7, pages 123-245 in The Standard Edition of the Complete Psychological Works of Sigmund Freud. London: Hogarth; New York: Macmillan. → First published in German.

Freund, K.; and Pinkava, V. 1961 Homosexuality in Man and Its Association With Parental Relationships. Review of Czechoslovak Medicine 7:32-40.

Goffman, Erving 1963 Stigma: Notes on the Management of Spoiled Identity. Englewood Cliffs, N.J.: Prentice-Hall.

Hooker, Evelyn (1957) 1963 The Adjustment of the Male Overt Homosexual. Pages 141-161 in Hendrik M. Ruitenbeek (editor), The Problem of Homosexuality in Modern Society. New York: Dutton.

Hooker, Evelyn 1965a Male Homosexuals and Their “Worlds.” Pages 83-107 in Judd Marmor (editor), Sexual Inversion: The Multiple Roots of Homosexuality. New York: Basic Books.

Hooker, Evelyn 1965 & An Empirical Study of Some Relations Between Sexual Patterns and Gender Identity in Male Homosexuals. Pages 24-52 in John Money (editor), Sex Research: New Developments. New York: Holt.

Huffman, Arthur V. 1960 Sex Deviation in a Prison Community. Journal of Social Therapy 6:170-181.

Indiana University, Institute for Sex Research 1965 Sex Offenders: An Analysis of Types, by Paul H. Gebhard, J. H. Gagnon, W. B. Pomeroy, and C. V. Christenson. New York: Harper.

Kinsey, Alfred C. et al. 1948 Sexual Behavior in the Human Male. Philadelphia: Saunders.

Kinsey, Alfred C. et al. 1953 Sexual Behavior in the Human Female. Philadelphia: Saunders.

Krafft-Ebing, Richard von (1886) 1965 Psychopathia sexualis. New York: Putnam. → First unexpurgated edition in English; original in Latin.

Lemert, Edwin M. 1951 Social Pathology: A Systematic Approach to the Theory of Sociopathic Behavior. New York: McGraw-Hill.

Leznoff, Maurice; and Westley, William A. (1956) 1963 The Homosexual Community. Pages 162-174 in Hendrik M. Ruitenbeek (editor), The Problem of Homosexuality in Modern Society. New York: Dutton.

Lindzey, Gardner 1965 Seer Versus Sign. Journal of Experimental Research in Personality 1:17-27.

Mahmor, Judd (editor) 1965 Sexual Inversion: The Multiple Roots of Homosexuality. New York: Basic Books.

Opler, Marvin K. 1965 Anthropological and Cross-cultural Aspects of Homosexuality. Pages 108-123 in Judd Marmor (editor), Sexual Inversion: The Multiple Roots of Homosexuality. New York: Basic Books.

Pare, C. M. B. 1965 Etiology of Homosexuality: Genetic and Chromosomal Aspects. Pages 70-80 in Judd Marmor (editor), Sexual Inversion: The Multiple Roots of Homosexuality. New York: Basic Books.

Perloff, William H. 1965 Hormones and Homosexuality. Pages 44-69 in Judd Marmor (editor), Sexual Inversion: The Multiple Roots of Homosexuality. New York: Basic Books.

Reiss, Albert J. JR. (1961) 1963 The Social Integration of Queers and Peers. Pages 249-278 in Hendrik M. Ruitenbeek (editor), The Problem of Homosexuality in Modern Society. New York: Dutton.

Romm, May E. 1965 Sexuality and Homosexuality in Women. Pages 282-301 in Judd Marmor (editor), Sexual Inversion: The Multiple Roots of Homosexuality. New York: Basic Books.

Ruitenbeek, Hendrik M. (editor) 1963 The Problem of Homosexuality in Modern Society. New York: Dutton.

[Schofield, Michael] 1960 Minority: A Report on the Life of the Male Homosexual in Great Britain, by Gordon Westwood [pseud.]. London: Longmans.

Schofield, Michael 1965 Sociological Aspects of Homosexuality. Boston: Little.

Schur, Edwin M. 1965 Crimes Without Victims. Englewood Cliffs, N.J.: Prentice-Hall. → See especially pages 1-11 and 67-119.

Society of Medical Psychoanalysts 1962 Homosexuality: A Psychoanalytic Study, by Irving Bieber et al. New York: Basic Books.

Sullivan, Harry Stack 1953 The Interpersonal Theory of Psychiatry. Edited by Helen Swick Perry and Mary Ladd Gawel. New York: Norton.

Sykes, Gresham M. 1958 The Society of Captives: A Study of a Maximum Security Prison. Princeton Univ. Press.

Tehman, Lewis M.; and Miles, Catherine C. 1936 Sex and Personality: Studies in Masculinity and Femininity. New York: McGraw-Hill.

Thompson, Clara (1947) 1963 Changing Concepts of Homosexuality in Psychoanalysis. Pages 40–51 in Hendrik M. Ruitenbeek (editor), The Problem of Homosexuality in Modern Society. New York: Dutton.

Ward, David A.; and Kassebeaum, Gene G. 1965 Women’s Prison: Sex and Social Structure. Chicago: Aldine.

Wilbur, Cornelia B. 1965 Clinical Aspects of Female Homosexuality. Pages 268-281 in Judd Marmor (editor), Sexual Inversion: The Multiple Roots of Homosexuality. New York: Basic Books.