October 29, 2016

"When linguistic analysis goes horribly wrong: no, Donald Trump doesn't 'talk like a woman.'"

Fighting back against that article we were talking about here the other day, the Columbia linguistics professor John McWhorter writes:
[I]t has been shown that women and men's speech does differ according to almost confoundingly particular attributes....The question is what it all means. It isn't an accident that it’s so hard to wrap your head around the idea of "the" and "below" as "guy talk."

One clear problem here is that femaleness is not the only heading the traits that Jones has identified fall under... A linguist recognizes all of these traits as more typical of casual, spoken language as opposed to formal, written language.

In speech, we are personal ("I"). We use a relatively basic vocabulary and we often grope for words beyond it, resorting to catch-all terms like "Whatchamacallit" and "that thing." In running speech, which is most of how we use language day to day, we are concerned with the immediate context rather than crafting abstractions about the broader world beyond us. Articles like "the" and "a" help us describe things — "a" for new things versus "the" for that which we already know. Prepositions are an especially odd aspect of Jones’s findings, but prepositions are part of placing new things in time and space.
Are you telling me that the linguistics research that extracted these male/female differences did not control the choice of texts so that the speakers were in comparable situations motivated by the same purposes?! Did they compare women's casual spoken word to men's formally crafted writings?
No. That would be insanely unscientific, but it's possible that women in the same situation would choose to speak in a more casual conversational style. So if linguists compare candidates' political speeches, they might find a female stereotype that's different from a male stereotype. As McWhorter puts it:
That in public speeches women take it somewhat more personal than men — although we are talking just tendencies here, not absolutes — is indeed news, and lends itself to assorted interpretations. However, it is this personal aspect of speech that Trump appears to model. That is, compared to average people presenting themselves in public, Trump is a highly personal speaker. Decidedly low on his list are crafting abstractions beyond everyday experience, fashioning new ideas, or stepping beyond the self.
And yet that's always what I took "talking like a woman" to mean. Women generally lean toward more casual, conversational speech. Not all women, and some men might do it too. It's a little inflammatory to refer to that as "talking like a woman," but it's an interesting tease.
The irony is that especially in our come-as-you-are times when formality is associated with inauthenticity (something Mitt Romney was hobbled by), this aspect of Trump-talk has much to do with his appeal. He talks like your friend on the barstool — exactly like him. If there's anything Trump is incapable of, it's artifice.
Yeah. Talking like a woman feels confidential and gets you involved and connected, and that works for men too if they can do it.

Why don't more men do it? Maybe they're afraid of seeming womanly and they lean toward performing stereotypical masculinity. Much of the time Trump himself is performing stereotypical masculinity. He does that openly and to the point where many people conclude that he's unusually masculine — blustering, frowning, making tough-sounding proposals. But maybe he gets away with that because it's balanced with femininity in ways that come in under the radar. We don't perceive him as effeminate, but the masculinity gets past our defenses because it is leavened with femininity. That's what I think.

But McWhorter says:
Of course, just as the idea that Trump talks "like a woman" seems ludicrous, the idea that women talk like Trump would seem to border on insult. 
That's why the statistical analysis was so interesting. It was counterintuitive. And it had great potential to rile up Trump haters. But to me, it was important because it helped explain the mystery of Trump's success and resiliency in the face of devastating attacks.

To McWhorter, Trump is just woefully unprepared and therefore using simple speech and repeating himself a lot. That happens to resemble what women do, but of course, women have different reasons for producing the statistical data that linguists have recorded.

I don't expect academics to explain female speech with cruel judgments about how ignorant and unprepared we are. I expect academics to explain the evidence they discover with ideas that compliment women: Women seek to build relationships and to exhibit empathy or whatever seems like a compliment and fits the data.

Look how McWhorter openly rejected insulting women. It can't be that Donald Trump talks like a woman, because to say that would be to insult women. Another explanation is required.

I'm still interested in the subject When linguistic analysis goes horribly wrong, but there are many ways to go wrong....

32 comments:

traditionalguy said...

Damn. Even linguistics is being rigged to defeat Trump.

Bob Boyd said...

He talks just like a woman
Combs his locks just like a woman
But he tweets just like a little girl

Laslo Spatula said...

"[I]t has been shown that women and men's speech does differ according to almost confoundingly particular attributes...."

Probably the best way to evaluate this is by examining the Sexting of men and women.

A common subject, where we can judge the dialogue on an even playing field.

Start with Carlos Danger and the fifteen-year-old girl. Obviously.

I am Laslo.

Achilles said...

Could it be that linguistics doesn't draw the brightest bulbs and "research" in this field is about as useful as sociology or psychology?

rhhardin said...

Men choose to be stereotypically male. Who knew.

There are apparent exceptions when they try to pick up a feminist.

rhhardin said...

I've noticed no difference in men's and women's morse code.

Only a small sample of women, three cases I think.

Paddy O said...

He talks like a car salesman. Says what needs saying to make the sale.

Is that a controversial concept?

Meanwhile, Hillary talks like an out-of-touch grandma who is trying to relate to someone else's grandchildren.

rhhardin said...

Women's morse code example

599 ERI

Guy's morse code example, same contest (New York QSO party)

599 ALB

traditionalguy said...

We are watching the expansion of a little creepy Trump into a popular figure who as easily relates to men and to women and to blacks and to whites and to Evangelicals and to Catholics and to Southerners and to Northerners and to Hindus and to Russians.

The trick is to keep up with the changes that are a happening. This writer is stuck on stupid.

Trump still has his problem relating to Jihdist Muslims and to Mormons. But that is because his style exposes vulnerable con-jobs where ever they are found.

Cacimbo Cacimbo said...

@BobBoyd Don't you mean he tweets just like a seer:

Trump tweet @ 10:50 AM - 3 Aug 2015

It came out that Huma Abedin knows all about Hillary’s private illegal emails. Huma’s PR husband, Anthony Weiner, will tell the world.

holdfast said...

Is it just me or is linguistics pretty much just whatever its "practitioners" want it to be.

Chomsky has somehow used it for decades as a cover for promoting communism and anti-Americanism.

rhhardin said...

Talk like an Egyptian.

madAsHell said...

As Sofia Vergara said, "What's wrong with having a dick in your mouth?"

rhhardin said...

I once interpreted what two women were really saying in a conversation between a librarian and a manager, into male direct speech. This was tolerated but not appreciated by the two women.

Unknown said...

"Are you telling me that the linguistics research that extracted these male/female differences did not control the choice of texts so that the speakers were in comparable situations motivated by the same purposes?"

It's not chemistry. It's not physics. It's closer to psychology, which is fraught with bias yet claims to be scientific.

chuck said...

Speech gender is easier to determine in Japanese where ther differences between male and female speech is far more explicit. My step sister studied Japanese, but got much of her practice waiting on Japanese men in Hawaii. When she finally got to Japan she was speaking male Japanese.

Terry said...

holdfast said...
Is it just me or is linguistics pretty much just whatever its "practitioners" want it to be.
Chomsky has somehow used it for decades as a cover for promoting communism and anti-Americanism.
10/29/16, 11:03 AM


It struck me the other day that what Chomsky had done for linguistics was to prevent it from saying anything definitive about language. All you can do is discuss it endlessly. Chomsky has done the equivalent of turning astrophysics back into astrology.

HT said...

To McWhorter, Trump is just woefully unprepared and therefore using simple speech and repeating himself a lot.

I agree wholeheartedly! - though I stick by my impression that he can also be engaging.

That happens to resemble what women do

That's what you say.

n.n said...

Trump is male, not female, not confused. There is no insult intended or implied.

They are patronizing the transgender/homosexual, crossover, etc. community. They need to stop.

mtp said...

What most people without a lot of math education, which is to say most people, don't understand is that all social science is just complete and utter nonsense dressed up in formal looking document formats.

Fen said...

Ah cool, a "bad science" tag.

Hat tip. I'm tired of all these bogus "studies" being treated as Science.

Sebastian said...

"Talking like a woman feels confidential and gets you involved and connected, and that works for men too if they can do it. Why don't more men do it? Maybe they're afraid of seeming womanly and they lean toward performing stereotypical masculinity." No. Because "feeling confidential" and "getting involved and connected" is usually irrelevant to fixing the leaky pipe or making the right play or understanding the balance sheet or writing the right code or (shudder) tearing apart the latest SCOTUS absurdity.

HoodlumDoodlum said...

Linguistic differences btw women & men? Impossible! There are no differences, and anyway gender is a social construct so all that could be studied would be the effect of social sexism.

Christy said...

Interpreting McWhorter, I conclude Bill talks like a woman, and Hill, not so much.

Speaking of unprepared, I've always thought Obama spoke as someone who sat through a lot of social science 101 classes, but never studied in detail or reflected on what it all meant. He was never deeply educated. Everything out of his mouth is incredibly shallow, just the kinds of BS I say when I don't know what I'm talking about.

Andrea said...

A few points: Sedivy's source is a graduate student working toward a degree in political psychology. I am a linguist, and I've always wondered what background knowledge researchers in other fields have when they wander over to linguistics. Maybe they've done their homework and know the issues and how to handle them--exactly along "the did you control for x,y,z?" But who knows? Maybe one day I'll go and ask.

At the same time, I never claim my work is scientific, and I don't believe linguistics should want to be considered a science. I empirically observe what people do with language in everyday life and look for patterns worth talking about. But just because my work is empirical doesn't mean that it's science.

From my perspective, linguistics is a thing like art or music or history. We aren't just industrious ants. We can reflect on ourselves, and studying linguistics is part of understanding ourselves. This is why linguistics is frequently found in humanities departments next to philosophy.

That said, the student's counter-intuitive result is interesting whether one leans her way or McWhorter's. More interesting would be to see the political speeches she made up that were "feminine" and "masculine" for the experiment she designed.

Random replies--
Within the field, Chomsky the linguist and Chomsky the political mad man are not conflated. That said, Chomsky and his disciples have had a stranglehold on the direction of linguistics for decades. This probably contributes to linguistics research being done under the disguise of psychology, anthropology, sociology, language teaching, etc.


As to the usefulness of linguistics as a field of study--
On the tech side, there is that artificial intelligence, Turing test, speech recognition, Siri thing. In the health sciences, there's speech pathology/therapy and Dr.-patient communication. Politically and more, there are issues in translation and interpretation. I had a prof who was an interpreter during the START missile negotiations. Probably a good thing that she was trained well.

Michael K said...

"This is why linguistics is frequently found in humanities departments next to philosophy."

It is actually useful in other areas as study of Indo European language does. Archeology and genetics are getting into this and linguistics could play a role i they get rid of Chomsky, et al.

I have read every book of Stephen Pinker, for example. I think he is very useful.

Char Char Binks said...

"In speech... we often grope for words..."

Can Trump be forgiven a little groping? Can we distinguish between him saying "their rapists" and "they're rapists"? Is he responsible for what we hear?

Eric the Fruit Bat said...

It's funny when I say "horribly wrong" maybe because I know I don't mean it.

whswhs said...

As Bob Dylan did not sing,

You talk just like a woman. . . .

XRay said...

Excellent post by Andrea at 2:00 PM. Thank you.

T2 D said...

If Donald really talks like a woman, that is a good thing. We have at least one candidate who talks like a woman. The other candidate brays like a donkey.

Inkling said...

Quote:"Are you telling me that the linguistics research that extracted these male/female differences did not control the choice of texts so that the speakers were in comparable situations motivated by the same purposes?!"

Alas. all too much academic research, particularly in the humanities and social sciences doesn't "control" for anything but the opportunity for achieving tenure and advancing careers. The more outrageous the discovery or the more it fits with liberal dogmas, the better. Carefulness, accuracy and truth are irrelevant.

Careerism has made science corrupt.