Anne Fausto-Sterling's books are informative and fascinating. She writes in an exploratory, non-dogmatic way that I really appreciate. She is hard to pin down and I (often) like authors who are hard to pin down. But one argument she seems to make in her books does not convince me much -- the argument that sex must be socially constructed, based on there being intersex individuals who wind up "assigned" to a sex in a social fashion.
About 1.7% of people are born with some sort of an intersex condition, she says. In these cases, decisions have to be made about whether the child will be brought up male, female, both, or neither. These decisions are typically made in light of cultural understandings of what is important in males and females. Therefore...what? Therefore, all sex categorizations are "socially constructed"?
Surely that doesn't follow. Why shouldn't we simply construe intersex individuals as borderline cases? There are clear male babies and clear female babies, and there are also individuals who fall in between. This is so in all sorts of other domains. There are clear chairs and clear couches, but also pieces of furniture that fall in between. There are clear trees and clear bushes, but plants that fall in between. Borderline cases can just remain borderline, unless there's some particular reason to categorize them. Maybe the furniture store has a chair room and a couch room, so we simply must put a chair/couch in one or the other. If we do that on some basis, such as which room has more space available, we don't have to think that has any general relevance to what makes chairs chairs or couches couches. Likewise, even if the sexes of intersex people are "socially constructed," if doesn't follow the sexes of clear cases are socially constructed.
And then, should we even embrace the social construction of sex in the intermediate cases? If you take "social construction" very literally, it seems to suggest we leave it up to society--the community, the state, the doctors, the family. If the community says it takes a penis to be classed with males, then so be it. If the community says it takes two X chromosomes to be classed with females, then so be it. But that's a terrible way to "assign" sexes to intersex babies. Fausto-Sterling actually advocates intersex children being tentatively (and non-surgically) assigned to a sex but later making their own choices based on how they see themselves. These kids will come to see themselves as male or female in a cultural context, so there is a social element there, but the child's self-perceptions have an internal component too, as I think she recognizes. If the child's eventual self-perceptions are given lots of weight, the sex classification of intersex children is at most partly "socially constructed."
As the chair/couch example shows, self-perception could be relevant to categorizing intersex people, but not relevant to clear-case males being male and clear-case females being female. The way borderline cases are dealt with does not necessarily have anything to do with how clear cases are classified. But perhaps that's merely a logical point: in principle, self-perception doesn't have to be relevant to clear males being male or clear females being female. But you might think it is relevant, even if it doesn't have to be. Clear males can come to have a sense of being female and clear females a sense of being male. If we do respect these self-perceptions for intersex individuals, then maybe self-perceptions should also take precedence when sorting supposedly clear cases into male and female categories. All maleness and femaleness would be defined in terms of self-perceptions, as opposed to self-perception entering the picture when other criteria aren't decisive.
That would be a win for the psychological nature of sex, not the social construction of sex. And it certainly would be a hard thing to embrace. It makes sense to think a truly intersex child has no sex until self-perceptions emerge, but some kids are born with a definite sex. Coming to see yourself as having a sex different from your natal sex is difficult for transgender kids precisely because there is (usually) a natal sex.
In any event, I really don't see at all how intersex children provide much support for the claim that all sex classifications are socially constructed. That seems to be the idea in Fausto-Sterling's work (again, she is hard to pin down), and she's had a lot of influence. But I don't see how this reasoning is supposed to work.