In areas of cultural analysis, terms are often read rhetorically, meaning as merely social actions with a persuasive or claim-making intent. E.g., to say "I know that ball will go in the net!!" isn't a claim involving actual knowledge, but something like a rhetorical act which appeals to the literal meaning of "know" in, say, something like an ironic/exaggerated/fabricated/social way. If one analysed the term "know" as-if this rhetorical context were its primary meaning, then one would conclude that all knowledge is merely a social presentation, knowing itself is no real thing in the world, only a game of making claims.
This is a deeply implausible primary meaning of "know", because it makes inexplicable why anyone would claim to know (ie., why would playing this game have any persuasive force?). It only makes sense if a literal meaning is available in which it is possible, and indeed quite common, to actually know things in an ordinary way. Then claiming to know, and being ironic/etc., makes sense.
It's no coincidence then that from this fields of analysis, in which any term whatsoever can take this merely rhetorical meaning, are terms like "man" given such a reading. However, the claim that this constitutes the only or even primary meaning of "man" is quite implausible. Since in the vast majority of cases, e.g., in biology, law, science, medicine etc. the authors are not taking "man" to be a kind of social rhetorical assertion. For example, pension ages differ in law across men/women -- its implausible to suppose that this law concerns itself with merely rhetorical actions of individuals insofar as they make claims to be men/women.
Indeed, as above, unless there is a literal meaning of "man" the social act of claiming to be one in a variety of non-literal contexts becomes meaningless. Consider eg., "I'm not really a man, I'm just playing one on stage with a lot of makeup". Here I'd suppose "man" has to have a literal meaning of having the characteristics of (human, adult) males in order for social claims of the sort, "I am a man!" to make any sense.
The very contemporary move to re-read these terms as primarily rhetorical claims might make some sense from a political/cultural perspective; but its quite important to remember that this cannot be their literal meaning -- or else a vast amount of derivative rhetorical meaning, and indeed existing law/textbooks/discourse/etc. becomes meaningless.
Two interpretations:
1) It has a literal meaning: the speaker has the characteristics of adult human males.
2) It is has a social-asserting meaning: the speaker wants to be perceived as "the kind of person who makes those assertions"
For asserting-meanings like (2) to make sense, literal meanings (1) have to be available.
Consider, "I'm not a crook!"
For it to make sense to claim not to be a crook, it has to be literally possible not to be one (ie., to be innocent of crimes) independent of claiming to be one.
Many today want (2) to be the "primary" meaning of terms like man/woman, ie., we are told that we should always read them as social assertions. But we have to keep in mind that literal meanings (1) have to exist in order for (2) to even make any sense -- not least that in almost all historical cases (most of the law, medicine, etc.), the literal meaning is the intended meaning.
To say "I am a man!" cannot always mean, "I wish to be seen as having the characteristics of adult males". Just as saying "i am not a crook" cannot always mean, "i wish to be seen as innocent".
A person saying, "I am not a crook" is hoping to persuade the audience that they are innocent. They are "borrowing" from the literal meaning, hoping to persuade the audience that are innocent -- even if they are not.
People saying "I am a man" who are not literally men are likewise borrowing from "what would make this claim literally true" when the say it. If they arent, it wouldnt make any sense to say. The audience has to already know what men literally are before a person claims to be one.