Nov 192013

stylo_logo.pngI was wandering around the Internet when suddenly an interesting approach to gender and (assumptions regarding) writing style appeared:

I have to admit that I thought it was a girl too based on the exaggeration and wording…”as if… universe.” Even the introducer sounded like a female because based on my original assumption that the submitter was a girl, I assumed that a man wouldn’t say, “smack some sense into”. Gosh! I had no idea that make such wild subconscious assumptions! actually, thank you for pointing that out. And I agree with the other person, we need a gender neutral pronoun. I usually employ ‘they’ even in the singular sometimes.

Signed “Kim,” a wonderfully ambiguous name that lets us speculate about the author’s own gender.

Sure enough, I decided to google “smack some sense into” and the result of this informal survey is that the first two pages are mostly from female authors. (No, not entirely. No, this is not the main point here; keep reading.)

Thus, the question I ask in three parts.

(1) Can this even be done in contemporary English, where we have a native understanding of the language and an almost infinite source of training samples? Can writing style be measured by an algorithm that has a better than a 50/50 success rate at guessing gender, preferrably better than a 80/20 success rate? Provisionally, and with a large enough sample, I would guess that the answer is yes.

(2) Which extant authors of antiquity are known to have been, or thought to have been, female? Is there enough material in Greek or Latin to attempt to train a statistical procedure for guessing the gender of the author of an ancient text?

(3) Can this be applied to some of the texts of antiquity from Christians as a point for consideration in the discussion that arises regarding their authorship?

If you don’t think this subject is important, you may say so, but don’t think it doesn’t matter or isn’t of interest to some people. If you don’t think the method could have any success or are skeptical, don’t worry, you have every right to be skeptical without any kind of demonstration. If you think you might have a lead regarding any of these three specific questions, that would most definitely pique my interest.

PS – Maybe I should have just done another Google search? The terms stylometry gender pull up some existing research.

  One Response to “A Stylometry of Gender?”

  1. Hi Peter

    Going back to “Luke” being a woman, I have been told gLuke & Acts could have been written by a male author who would be extremely feminist, for a variety of reasons.
    (Reluctantly) maybe, but there is one thing in gLuke which betrays a woman’s hand:
    Lk 21:16-18 “You will be betrayed even by parents and brothers, relatives and friends; and they will put some of you to death. And you will be hated by all for My name’s sake. But not a hair of your head shall be lost [last sentence only in gLuke!].”

    Let’s compare this with:
    1 Cor 11:6b ” … if it is a disgrace for a woman to have her hair cut or shaved off, she should cover her head.”
    1 Cor 11:15a “But if a woman have long hair, it is a glory to her …”

    In case of persecution (resulting in death for some), “Luke” was concerned about hair being cut or shaved off!

    Could that comment come from a man?
    I do not think so: many men were bold or bolding, so loosing hair was not a big issue.

    I do not think we can dream up of a procedure. But an other abundance of feminist items in one gospel, as compared with the other ones, plus the case I just explained, would close the book for me.

    That webpage of mine can be consulted. It shows a large amount of data converging to the fact “Luke” was a woman (and also a Roman citizen from Philippi, a Roman colony, where a Christian church was created first among women, according to Acts). If it is not enough, what is?

    Cordially, Bernard

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>