Research by John Money and others over the past few decades has shown that when a child is born hermaphroditic (of indeterminate sex) but is decisively identified as one sex or the other by doctors and parents from a very early age, that child will usually be fairly accepting of the assigned gender and the social role that accompanies it (Money and Ehrhardt, 1972). This fact has been quoted by many social learning theorists as support for their position that sex roles are acquired rather than inborn. If a child is genetically one sex but can successfully be reared as a number of the other, they say, this must indicate the overwhelming power of social learning experiences. It is important, however, to recall in this connection that the biological factor determining gender is not the chromosome pattern but the balance of hormones to which the body and brain is exposed from conception onwards. An infant whose external genitalia are ambiguous is clearly only part masculinized and therefore genuinely intermediate in terms of gender, regardless of chromosome pattern. It is understandable that in cases like these it is possible to push the individual’s identification towards one gender or the other with a combination of surgical and hormonal treatment and social persuasion, so that they are unlikely to dispute their assigned gender. This tells us nothing about the difficulties that would be encountered in trying to cross-socialize an individual of unambiguous sex without the aid of surgery or hormones. Money and his colleagues now recognize this distinction and no longer maintain the radical environmentalist viewpoint that is often quoted from their earlier papers.

In any case Money’s early conclusions related specifically to gender identity, not the broader aspects of masculinity and femininity which are under partly separate brain control, being laid down in different area of the brain and at different phases of pre-natal development (Ellis and Ames, 1987). Many of the androgenital cases, for example, identified themselves as women happily enough but showed distinct ‘tomboyism’ in their pattern of interests. They enjoyed ‘rough-and-tumble’ activities, preferred wearing trousers to dresses and were uninterested in marriage and motherhood.

Since Money’s early work on hermaphroditism in humans, a fascinating new condition has come to light as a result of its prevalence within a particular mountain district of the Dominican Republic. Male pseudo-hermaphroditism (Imperato-McGinley et al., 1974) or machihembra (‘man-woman’ as the locals call it) is an inherited enzyme deficiency which causes genetic males to develop as females until puberty, at which time androgen production is suddenly increased and the individual turns into a male (complete with penis, descended testicles, deep voice, facial and body hair and well-developed muscles). The interesting thing about these individuals is that, although raised as girls, they have little difficulty in adjusting to a male identity after their bodily conversion at puberty. They develop male mannerisms and attitudes and are sexually attracted to females just like normal men. Perhaps most interesting of all is the discovery that they often begin to think of themselves as boys at the age of five or six, well before any physical changes have begun to take place.

This may be regarded as one of the most direct tests of the relative power of society and hormones in determining gender identity and sex-role characteristics, and the hormones appear victorious.