Your gender and your writing

[Pure]

He and she: What's the real difference?

According to a team of computer scientists, we give away our gender in our writing style

By Clive Thompson, 7/6/2003 {Boston Globe}

http://216.239.51.104/search?q=cach...shtml+gender+computer+writing+&hl=en&ie=UTF-8[/url]

MAGINE, FOR A SECOND, that no byline is attached to this article. Judging by the words alone, can you figure out if I am a man or a woman?



Moshe Koppel can. This summer, a group of computer scientists-including Koppel, a professor at Israeli's Bar-Ilan University-are publishing two papers in which they describe the successful results of a gender-detection experiment. The scholars have developed a computer algorithm that can examine an anonymous text and determine, with accuracy rates of better than 80 percent, whether the author is male or female. For centuries, linguists and cultural pundits have argued heatedly about whether men and women communicate differently. But Koppel's group is the first to create an actual prediction machine.

A rather controversial one, too. When the group submitted its first paper to the prestigious journal Proceedings of the National Academy of Sciences, the referees rejected it ''on ideological grounds,'' Koppel maintains. ''They said, `Hey, what do you mean? You're trying to make some claim about men and women being different, and we don't know if that's true. That's just the kind of thing that people are saying in order to oppress women!' And I said `Hey-I'm just reporting the numbers.'''

When they submitted their papers to other journals, the group made a significant tweak. One of the coauthors, Anat Shimoni, added her middle name ''Rachel'' to her byline, to make sure reviewers knew one member of the group was female. (The third scientist is a man, Shlomo Argamon.) The papers were accepted by the journals Literary and Linguistic Computing and Text, and are appearing over the next few months. Koppel says they haven't faced any further accusations of antifeminism.

The odd thing is that the language differences the researchers discovered would seem, at first blush, to be rather benign. They pertain not to complex, ''important'' words, but to the seemingly quotidian parts of speech: the ifs, ands, and buts.

For example, Koppel's group found that the single biggest difference is that women are far more likely than men to use personal pronouns-''I'', ''you'', ''she'', ''myself'', or ''yourself'' and the like. Men, in contrast, are more likely to use determiners-''a,'' ''the,'' ''that,'' and ''these''-as well as cardinal numbers and quantifiers like ''more'' or ''some.'' As one of the papers published by Koppel's group notes, men are also more likely to use ''post-head noun modification with an of phrase''-phrases like ''garden of roses.''
====

Koppel papers are at
http://www.cs.biu.ac.il/~koppel/male-female-llc-final.pdf

http://www.cs.biu.ac.il/~koppel/male-female-text-final.pdf


Data (books) and features lists are at
http://sukka.jct.ac.il/~argamon/gender-style/

 

[Lauren Hynde]

No offense intended, but what a monumental crap of bull.

English isn't my mother language and I don't use it extensively in my everyday life, except in writing. It is, however, a language I'm constantly absorving via television, cinema, internet, and more importantly, books.

I am certain that over 80% of what I read is written by men. You can be certain that factor will be much more determinant to my writing style, to my thinking process, in fact, than any variable present in Moshe Koppel's study.

Your education, your geographic location, your cultural references, a myriad of other factors will differentiate your writing from another woman's whose outlook on life has nothing to do on yours and bring it closer to a man who shares your background.

 

[Chicklet]

Hmm...if people can tell my gender through my writing why do I still get emails from women saying they want to suck my cock?

 

[Guest]

Very good points Lauren. I'll repeat myself from a seemingly long ago thread: Language is inherently sexist, and I do not merely speak of exclusivity in pronouns. If you think about what language is, the human creation and necessity of it, and its use in self identity, then the fact that it has been dominated, created, shaped, ruled, etc., by men for centuries might give one pause for thought about a silly study as the above. Women are barely at the first stage of claiming their own voice(s). I daresay it will be centuries before language "evolves" in equality. I've been reading a bit over five decades and am proud *and* fortunate to have the voice I have. A woman might have come up with a better study than the above, IF she had the time, backing (including filthy lucre) and credentials.

Personal note: Good to see you Pure, I've missed you,
sincerely,
Perdita

 

[Svenskaflicka]

I seriously doubt that there is any biological difference between men and women, apart for a few organs.
I think that the reason why men use one type of language and like sports and mechanics, while women use another type of language, and get stuck with cooking and cleaning while their men stare at the damned NBA-games, is Society's Fault.

We're raised, not only by our parents, but by the whole society, to adopt certain gender roles. If we obey, we're accepted. If we try to rebel and be more individual, people regard us as freaks.

 

[Guest]

Society = Patriarchy (still today)

 

[MathGirl]

Semites, Schemites

Originally posted by Pure This summer, a group of computer scientists-including Koppel, a professor at Israeli's Bar-Ilan University-are publishing two papers in which they describe the successful results of a gender-detection experiment.

Dear Pure,
Well, you know how those Jews are.
MG

 

[Pure]

LH said

//I am certain that over 80% of what I read is written by men. You can be certain that factor will be much more determinant to my writing style, to my thinking process, in fact, than any variable present in Moshe Koppel's study.//

Why can I be certain of that? What's your evidence?

I've posted the links to the papers, so perhaps it would be good have a look at the method and variables before drawing conclusions.

J.

 

[Pure]

Svenskaflicka said,

I seriously doubt that there is any biological difference between men and women, apart for a few organs.

Agreed, provided the brain is included as an organ.

I think that the reason why men use one type of language and like sports and mechanics, while women use another type of language, and get stuck with cooking and cleaning while their men stare at the damned NBA-games, is Society's Fault.

The reasons for the differences in language, assuming there are some, can be many, I agree, and of the type you mention. Lauren also mentioned reasons. The papers, however only addressed outcomes: hundreds of papers and books by males and females, and the question of differences in syntax (not subject matter).

 

[Why]

Hmmmmmm... never been one for putting my gender up on forums. Always tried to be asexual as possible. Try to do the same with my writing want it judged for the content, not because I have balls or breasts.

It would be interesting to know, which "sex" they pegged me as.

 

[Never]

Chicklet:
"Hmm...if people can tell my gender through my writing why do I still get emails from women saying they want to suck my cock?"

Human beings aren't computer programs designed to guess your gender based on your writing style.

Perdita:
"...the fact that it has been dominated, created, shaped, ruled, etc., by men for centuries might give one pause for thought about a silly study as the above."

High-class males have dominated written language, yes. However, the majority of the world's population has always been illiterate.


Pure,
I'd say the brain and the endocrine system.

I skimmed the paper you posted and it was quite interesting. I'm surprised the article didn't mention that the computer could determine the difference between fiction and non-fiction 98% of the time. I also thought it was interesting that if you 'taught' the program on only non-fiction, it's ability to distinguish between male and female fiction writing dropped to 50%

 

[Pure]

Hi Never,

I glad you gave the material a read. It's amazing how some people without reading papers can judge them--as 'crap' (LH) or as easily improved on by a woman (Perd). Indeed a woman was among the investigators.

You say,

//.I also thought it was interesting that if you 'taught' the program on only non-fiction, its ability to distinguish between male and female fiction writing dropped to 50%..

That's true; there are differences in markers, fiction and nonfiction, though the main study tried for a discrimination of both kinds of authorship, from one algorithm based on analyzing fiction and non fiction.

But one shouldn't forget that this does mean an inability to discriminate female written fiction based on analysis of the fiction sample.



The first paper says,

When we train only on fiction documents (thus using a substantially smaller training set), results of 36-fold crossvalidation (maintaining ten examples per fold) actually increase to 79.5%. Likewise, when training on non-fiction only, accuracy on non-fiction test documents increased to 82.6%. This is because, as we shall see in detail below, the frequencies of the critical distinguishing features are different in fiction and non-fiction. In fact, when training is performed on fiction (non-fiction) only, non-fiction (fiction) documents are categorized with barely more than 50% accuracy – no better than random!

 

[Pure]

Here are two excerpts of fiction--extreme cases--chosen by the authors to illustrate the differences. Their comments are at the end.


Saigon. Grey, Anthony
BY 1925 present-day Vietnam was divided into three parts under French colonial rule. The southern region embracing Saigon and the Mekong delta was the colony of Cochin-China; the
central area with its imperial capital at Hue was the protectorate of Annam; and the northern region, Tongking, was also a separate protectorate with its capital at Hanoi. The Annamese
emperor, Khai Dinh, in theory ruled the two northern regions from Hue with the benefit of French protection, while Cochin-China was governed directly from Paris but in effect all three
territories were ruled as colonies. Some backward tribes inhabited the remoter mountains and jungles but the main population was of the same race; today they are known as Vietnamese but then the outside world knew them as Annamites or Annamese. They had detached themselves from the torrent of peoples that in prehistory had poured out of China onto the countless islands of the Pacific and, settling the eastern coastal strip of the Indochina


Jerusalem the Golden. Drabble, Margaret.
Clara never failed to be astonished by the extraordinary felicity of her own name. She found it hard to trust herself to the mercy of fate, which had managed over the years to convert her
greatest shame into one of her greatest assets, and even after years of comparative security she was still prepared for, still half expecting the old gibes to be revived. But whenever she was
introduced, nothing greeted the amazing, all-revealing Clara but cries of “How delightful, how charming, how unusual, how fortunate,” and she could foresee a time when friends would
name their babies after her and refer back to her with pride as the original from which inspiration had first been drawn. Finally her confidence grew to such an extent that she was able
to explain that she had been christened not in the vanguard but in the extreme rearguard of
fashion, after a Wesleyan great-aunt, and that her mother had formed the notion not as an unusual and charming conceit but as a preconceived penance for her daughter, whose only
offences at that tender age were her existence and her sex.
[end Drabble excerpt] ]/i]


[Koppel et al.]
These passages illustrate in extreme fashion the fundamental differences borne out by our statistical findings. Grey opens his book with a recitation of facts; Drabble opens hers with her
protagonist's thoughts. Consequently, Drabble uses 17 singular feminine pronouns, while Grey uses only four animate pronouns altogether and all are plural. In his 161 words, Grey uses 46
proper or common nouns, while Drabble uses only 33 in 187 words. Grey uses four numbers, Drabble none. Grey uses the determiner the 18 times, Drabble only 9. Overall, one could easily
imagine Grey's introductory passage in a non-fiction work, while Drabble's passage is unmistakably fiction.
=====

 

[Guest]

Perdita: "...the fact that it has been dominated, created, shaped, ruled, etc., by men for centuries might give one pause for thought about a silly study as the above."

High-class males have dominated written language, yes. However, the majority of the world's population has always been illiterate.

Hi Never. Just to point out I wasn't referring only to the written word. Language rules the illiterate too, especially the illiterate and other disadvantaged people.

All good points, and I did read the papers, but can't comment at their level of thinking, and don't care to in the time required to do so thoughtfully. Outside new feminist and new literary criticism it's still too infrequent to find any scientific work/theories that take human patriarchal history into account.

Pure, a woman on the team doesn't mean anything.

Perdita

 

[Lauren Hynde]

LH said

//I am certain that over 80% of what I read is written by men. You can be certain that factor will be much more determinant to my writing style, to my thinking process, in fact, than any variable present in Moshe Koppel's study.//

Why can I be certain of that? What's your evidence?

I've posted the links to the papers, so perhaps it would be good have a look at the method and variables before drawing conclusions.

J.

I glad you gave the material a read. It's amazing how some people without reading papers can judge them--as 'crap' (LH) or as easily improved on by a woman (Perd). Indeed a woman was among the investigators.

Well, I did read the papers several days ago; this isn't the first time these links have been posted here in AH.

The problem I have with the study doesn't have anything to do with the numbers. 80% accuracy is a very disappointing result, if they're trying to prove anything. If I were to guess the accuracy of a human to do the same job, it wouldn't be an unreasonable number to assume.

The problem is the study is presented as if demonstrating an innate physiological distinction between the way men and women write. I know there is nothing in the published papers refering to this directly, but that is their intention and how they will be approached. We all know it.

What I was saying was that these results, as those two excerpts clearly show, steam from a difference in style. But the study doesn't even show that the majority of men tendencially write in a particular style and the majority of women in another, but simply reveals a tendency of some publishing companies chiefly between 1975 and 1993.

Does this, however, has anything to do with the gender of the authors? I sincerely doubt it. There are, if you will, "masculine" and "feminine" literatures. One is sharper, edgier, more factual or presented that way, more cerebral. The other is more appealing to the senses, more sentimental. Historically, one has been the exclusive hunting-ground of men. The other has been permitted to women.

The choice between one and the other, however, is very rational. There's no reason why I woman can't write in a more cerebral fashion, just like there is no reason for men to be incapable of writing sentimental pieces if they choose to do so. A majority of authors stick to their "gender literature", however, because that's what they learned. That's what is expected of them. And in art, especially in literature, there are far more sheep than wolves. Certainly more than what we were all taught to believe.

Think of George Eliot. Oh, I'm sorry, I mean Mary Ann Evans. Or else of George Sand, known to his friends as Aurore Dupin. Or A.S. Byatt, curiously enough, one of the "misclassified" authors by Moshe Koppel's study. Hey, there's no need to go that far: think of JK Rowling; Joanne was probably too big a name to fit the cover of a book boys would want to read.

I don't want you to mistaken my position with any feminist conception. I don't fit that profile. Men and women are different. I simply don't believe those differences include any innate higher nervous system functions, and this study doesn't come remotely close to begin proving it.

Again, given all the cultural programing, does an 80% accuracy rate sound that impressive?

 

[Pure]

Lauren said,


//Well, I did read the papers several days ago; this isn't the first time these links have been posted here in AH.//

Yes, it's all so boring and trite. Yes, I see the links were posted by slicktony a few days ago, but no serious discussion ensued, just the usual DirtySlut/MathGirl & co. shimmy and shake. There is no evidence any of the responding posters read the papers.

Lauren, you otherwise make a number of good points, but the ones below are quite odd.


The problem is the study is presented as if demonstrating an innate physiological distinction between the way men and women write.

What is your evidence that the authors believe in 'innate' physiological distinction.? or _write as if demonstrating_ such a distinction?

I know there is nothing in the published papers refering to this directly,

OH, I see, there's maybe no evidence, nothing in the papers to support this allegation. Or maybe it's 'indirect' reference?



but that is their intention and how they will be approached. We all know it.

We have no evidence from the paper. But 'we' know their intention. I guess you have some other sources of information about the authors' intentions?

Or is it just that you're psychic?

----
The authors claim that there are statistical patterns in syntax in a few hundred works examined, that allow 'automatic' (algorithmic)discrimination of author gender (in a certain body of fiction and non fictional modern Western writings) with fair or modest accuracy. (There are also 'misses' described rather openly, as in Byatt's and Ishiguru's cases.)

Do you agree or disagree with this?)

I'm sure you 'know' a number of their other positions, and I see you disagree with what you've imputed, but what say you on the main point?


Lauren said
There's no reason why I woman can't write in a more cerebral fashion, just like there is no reason for men to be incapable of writing sentimental pieces if they choose to do so. [

You say this as if someone or Koppel is disagreeing with you. See my closing remark below, which makes the same point.

======
As a very small time author I'm intrigued with the differences. Since I write often in the "I" form, and often from the female point of view, i'm eager to learn the marks of female narration (at this point in history, before the millennial age of equality).

J.

PS Doesn't the Drabble excerpt remind you a bit of Math Girl's fictional style?

 

[Lauren Hynde]

Pure-

First, let me say I reread my posts and realize I may have come off as overly agressive, which wasn't my intention. I apologize for that.


Your questions-

The object of this study was to find an automatic way of categorizing a written text according to style. They were trying to find elements that would allow a computer to separate "male"/cerebral literature from "female"/sentimental literature. There are practical applications for this technology, namely in internet and database searches.

The results of the study were presented, however, in probably no more than a successful publicity stunt (since we actually heard of its existence), as capable (well, 4 out of 5 times) of determining the gender of the author of a text. Can you see any correlation between this and the aforementioned object of the study? I can't, frankly.

Not only that, the only uses I can think of for a program whose purpose is to determine the gender of the author of a text would be to promote discrimination of one of the genders in a private online community, to find out if Homer's books weren't in fact written by his mother, or to determine the gender of the pilots of an invading intergalactic squad, provided they previously give us some dully categorized samples of their work. None of these purposes will matter much, given the 20% miss probability.

So, without any other discernable purpose, logic dictates that to show differences between the writings of men and women is in itself a purpose. And since it doesn't take a team of trained scientists to see that there are in fact historical differences between the type of books commonly written by men and the type of books commonly written by women, the only thing left is to find whether or not those differences are genetically determined.

The National Academy of Sciences' people knew this was going to be what people would be thinking about when reading the study; this was what perdita, Svenska, I, and in all probability the vast majority (yes, I'm guessing) of the people who read this thread and the study first thought about.

I said there's no reason why a woman can't write in a more cerebral fashion, just like there is no reason for men to be incapable of writing sentimental pieces if they choose to do so, as if Koppel is disagreeing with me, yes. He is disagreeing with me because the misclassifications are presented as program failures.

I started by saying this study was a monumental crap of bull and I stand by it (and shuddup with the smartass remarks ). The biggest evidence of this is that, as far as we can tell, they may have totally succeeded. Not in their publicized objective, but the original one. They may have found a way of automatically separate cerebral and sentimental styles. The 20% error may be the brave authors who decided to write what they damned pleased, instead of what they were expected to. They may have made a breakthrough, but either decided to sacrifice their objectives in exchange for some publicity, or got so tangled up in their side job they forgot about the main thing.

Your desire to find the characteristics of both genders to better write as either of them is admirable, and I respect you for that, Pure. But what you need is to find out how to inflate personality to your characters and to your writing, not yourself. You won't find any help in this study. It's not what it's about.

 

[Pure]

Lauren, you said,


The object of this [Koppel]study was to find an automatic way of categorizing a written text according to style. They were trying to find elements that would allow a computer to separate "male"/cerebral literature from "female"/sentimental literature. There are practical applications for this technology, namely in internet and database searches.

Later you say,

they may have totally succeeded. Not in their publicized objective, but the original one. They may have found a way of automatically separate cerebral and sentimental styles.

Koppels first paper has this title:

Automatically Categorizing Written Texts by Author Gender

Koppel early on says this about his object:

//The object of this paper is to explore the possibility of automatically classifying formal written texts
according to author gender. (p.2)//

Determination of author gender is an original, stated goal, here, in the title, also in the abstract.

On what basis do you lay aside the authors' clear words and impute an 'original objective' other than this?

The 'cerebral' and 'sentimental' characteristics were not assumed; the investigators looked at nonfiction and fiction; they selected for representation of gender, but not 'sentimentality' or cerebral-ness.


The results of the study were presented, however, in probably no more than a successful publicity stunt (since we actually heard of its existence), as capable (well, 4 out of 5 times) of determining the gender of the author of a text. Can you see any correlation between this and the aforementioned object of the study? I can't, frankly.

They were presented so (related to author gender), because that was the clearly stated object, beginning in the title!

No I don't see a proper connection between what you *claim* was the 'original object', and what you say, correctly, is a characteristic of the presentation.

As to your statement:
//Not only that, the only uses I can think of for a program whose purpose is to determine the gender of the author of a text would be to promote discrimination of one of the genders in a private online community [and etc.]//

Wildly paranoid, or 'unduly apprehensive'... hard to say. I'll let readers determine that. I don't see where the papers discuss 'use' of the program. I gave my use. Others are possible. Perhaps writers need to be aware of their syntax, and work with it.

It's a kinda shame to see your intelligence and education being laid aside because of some prejudiced reactions.
J.

 

[Never]

I don't think it's unusual to publish information on a program that's main use is to be interesting to publish.

 

[SlickTony]

Well, I did read the papers several days ago; this isn't the first time these links have been posted here in AH

Yes, I see the links were posted by slicktony a few days ago

Thank you, Lauren! Thank you, Pure! I'm glad to see people have decided to discuss this topic, even though it took somebody besides me bringing it up to do so.

I am reminded of the old joke about a guy who committed a crime and was sentenced to prison. The first time he came to the recreation room, he was puzzled to hear prisoners calling out numbers, and this would be followed by uproarious laughter from the other men.

This went on for several evenings, and finally he got up the nerve to ask someone what was going on.

"You see, we've all been here so long that we know all the jokes, and we've given them all numbers. We're just saving time by yelling out the numbers, because everyone knows what the jokes are."

The next evening, the jokes start, and the new guy, trying to enter into the spirit of the thing, yells out, "Number 75!"

Silence. Crestfallen, he asked one of his new friends what had gone wrong.

"Don't take it to heart, son," said an old-timer. "Some people can tell 'em and some people can't."

[Torn between a desire to continue reading, and maybe contributing to, the discussion, and going off somewhere to sulk.]

 

[Madame Manga]

Not having the time right now to read over all this research in detail--

Publishing companies often insist, I am told, that a 'feminine' book must have a feminine name on the cover for marketing purposes. What was the last time you saw a male name on the cover of a romance novel, for instance? If an author's real name is androgynous or not very fancy, she may well adopt an ornamental pseudonym just for the purpose of selling books. Even some well-known writers change their names when they switch genres--Nora Roberts writes murder mysteries as J.D. Robb. If anything too edgy or technical creeps into a romance manuscript, out it goes--it doesn't fit 'category'.

Vice versa for 'masculine' books, natch. Tom Clancy type techo-thrillers? Would "Red Storm Rising" have sold nearly as well if written by, say, 'Victorina Pendergrass'? ;-) Books like that are deliberately purged of thoughts and feelings in favor of numbers and hardware. Clancy happens to be peculiarly averse to personal relationships in his writing and so has been extremely successful in his niche, but similar books with too much 'mushy stuff' surely get that edited out somewhere in the process.

My not-well-illustrated point being, that which gets published is generally that which falls into easily salable categories: comfortable, familiar, sex-aligned in tune with the preconceptions of the buying public. Editors encourage writers to follow what has succeeded in the past, and so that is what everyone but the trailblazers ends up regurgitating. Publishing is not a business devoted to shattering stereotypes, just feeding on them. No wonder the commercial booklists seem to divide easily into masculine and feminine worlds.

MM

 

[Guest]

Madame Manga:

Just to say I always count on you for practical, no-nonsense posts.

admiringly, Perdita

(Which is not to say the posts above you are the opposite.)

 

[Pure]

MM,
Thanks for the backgrounder. I think you're saying that 'masculine' and 'feminine' styles' have a self perpetuating quality, based on expectations, both in publishers and readers (after all what would the publishers care if the readers didn't).

It would indeed be interesting to see if the Koppel program could detect those romance novels written by men with female pennames.

Keep in mind that the program looks at topic neutral things, some rather subtle. Would the men have mastered this, perhaps unconsciously. Consciously, of course, a man would have no trouble with the key idea of Koppel and predecessors: that the femal style is 'involved' and involves the reader. Hence the greater use of 'you' and 'not'

I have no problem with your main points. Further, even in HS, I imagine teachers see writings by young men and young women, and may be unconsciously 'rewarding' the following of usual patterns. The authors don't really get into causes, but I can see social ones are very important.

OTOH, as Lauren and others including the original critics have reminded us, the accuracy of 80% is far from perfect. The 'exceptions' might bear scrutiny, both textually and biographically. Why does AS Byatt NOT fit? Did she grow up with the odd job of writing catalog entries for a department store?

For my purposes, the lists of features are often NOT what I consciously look at, and it's interesting to see them, e.g., the 'garden of roses' construction N Prep N, with the Prep phrase being adjectival.

It's also to be noted that the often addressed topic of detecting male and female speech has proven to be a real mine field. (You know of the 'tag question' [alleged] marker?) Not so easy as thought, because of starting with preconceptions. Male and female written texts have NOT been really looked at, starting from a neutral basis, as Koppel did-- i.e., lists of hundreds of words and features that are topic neutral.

Regards,
J.

 

[Lauren Hynde]

Pure-

I would expect you to have read the papers before starting a discussion about them, but I see where I made my mistake.

If you had read the papers, you would know what I was talking about:

in 'Automatically Categorizing Written Texts by Author Gender' by Moshe Koppel, Shlomo Argamon, Anat Rachel Shimoni
[...]
Driven by the problem of Internet search, the text categorization literature – outside of the stylometric research community – has, with a few exceptions (Argamon-Engelson et al 1998, Wolters & Kirsten 1999), concerned categorization by topic rather than categorization by writing style. The problem considered in this paper concerns categorization by style...
[...]

Determination of author gender is an original, stated goal, here, in the title, also in the abstract.

On what basis do you lay aside the authors' clear words and impute an 'original objective' other than this?

The 'cerebral' and 'sentimental' characteristics were not assumed;

Do you believe that the title and abstract of a paper are written before or after the research? Before of after they decide which is the best angle to present the data? "They were presented that way because that's the title"? Does that sentence make any logical sense?

It is so easy to see through this paper it wouldn't even be "interesting to the public", if it wasn't for the joke potential.

The fact that Moshe Koppel's study looked for 'neutral' markers, as opposed to topic specific words, isn't the least bit surprising, since it wasn't their objective to separate text by topic, but by style. The results are a direct consequence of most men writing "masculine" books and most women writing "feminine" book, as I defined them in an earlier post, as Madame Manga did on hers, as publishing companies have been doing since the beginning of time.

The results the computer program returns have nothing to do with gender: they separate "masculine"-type books from "feminine"-type books. Any author of any gender can write a book and be misclassified; they only need to be flexible enough to write outside of their genre. Shift from a more cerebral writing style to a romance novel, or vice versa. Do you want to know why AS Byatt didn't fit? Did you read Possession?

As to your statement:
//Not only that, the only uses I can think of for a program whose purpose is to determine the gender of the author of a text would be to promote discrimination of one of the genders in a private online community [and etc.]//

Wildly paranoid, or 'unduly apprehensive'... hard to say. I'll let readers determine that. I don't see where the papers discuss 'use' of the program. I gave my use. Others are possible. Perhaps writers need to be aware of their syntax, and work with it.

How very interesting that you chose to replace the rest of my sentence with [and etc.]. You can try to distort what I said all you want, but my words are still up there for everyone to see. If you can't appreciate that the point I was trying to make was that there isn't any practical use for a program that discerns the gender of the author other than to use it to prove biological differences on the way men and women think, I can now see why you're expecting to learn the marks of female narration from this study. Good luck with it, I'm going to try to learn how to make a space-shuttle by making a list of their times of departure and average speed.

It's a kinda shame to see your intelligence and education being laid aside because of some prejudiced reactions.

Well, that's smart. I wasn't the one assuming I hadn't read the papers, nor was I assuming I was psychic, just because I chose to think. I apologize if I stepped out of my gender boundaries.

I am curious, though. What am I prejudiced against?

 

[Pure]

lh
//I apologize if I stepped out of my gender boundaries.//

that's ok, dear, just let me open this door for you.