Speaking as David Brooks is hard

David Brooks has an opinion piece in the New York Times called “Speaking as a White Male…”. It’s about identity politics and it features the usual headscratcher ideas that we have come to expect from the Times‘ opinion page, including this nonsense right here:

Brooks Opinion Speaking as a White Male - nonsense paragraph


Brooks’ column may be full of the stuff that only white dudes could ever think of, but I want to look at one particular thing that he says:

Now we are at a place where it is commonly assumed that your perceptions are something that come to you through your group, through your demographic identity. How many times have we all heard somebody rise up in conversation and say, “Speaking as a Latina. …” or “Speaking as a queer person. …” or “Speaking as a Jew. …”?

Brooks’ choice of words is telling (“rise up”, “Latina”, “queer”, and “Jew”), but I’m not going to get into that here (or just yet). I can’t remember hearing anybody rise up in conversation and say “Speaking as a(n) X”. So I thought I’d have a look at the SPOKEN section of the Corpus of Contemporary American English (COCA) to see if we can find this phrase. The data in the SPOKEN section of COCA comes from transcripts of news shows. Not all of it is entirely spontaneous, but it’s good enough for what we’re looking for. Here’s a search for “SPEAK as _at*” (that is: all forms of the lemma speak, followed by the word as, followed by an article):

COCA Spoken - SPEAK as _article_
Search in COCA for “SPEAK as _a*”

The first thing we can see is that speaking as a is the most common form of this construction, but that it seems to be trailing off in usage from the 1990s to 2017. On top of that, 63 hits are not that much (48 hits for speaking as a + 15 hits for speaking as an = 63). Let’s take a look at the two most common constructions.

Of those 63 hits, 35 fit the form that Brooks used. I’m excluding examples like “I’m not speaking as X” or “I was speaking as X” or when the speaking as X was put into the person’s mouth, such as in this example from NPR’s Wait Wait… Don’t Tell Me! Show:

Peter Sagal: And speaking as an esteemed historian, a Pulitzer Prize-winning historian, do in fact, in your scholarly opinion, the Yankees suck?

Doris Kearns Goodwin: Without a question.

Peter Sagal: Thank you.

Of the 35 hits that are similar to the form Brooks made up, most of the words that appear after speaking as a(n) are unique, so most of them appear only once. Some favorites of mine are

Speaking as a(n)

old-time criminal defense lawyer


reporter who missed that story

There are 2 hits each for Speaking as a followed by woman, individual, and mom. The words Latina, queer or Jew do not appear in the search results. There is, however, one example of speaking as a homosexual and one example of speaking as an Israeli dove. So two out of three ain’t bad?

What’s going on here, then? My guess is that Brooks inflated the number of times he has heard Speaking as a(n) X to suit his argument. It’s probable that he has heard it a couple of times, but he makes it seem like it’s an everyday thing. Of course, maybe just hearing speaking as a queer person once is one too many times for some people.

The scant results from the search could also indicate that the data in the corpus doesn’t include speech from enough people who identify as a Latina, queer or Jew. That is probably true (these people need to be represented more on our news shows), but I also think that people do not need to say speaking as a(n) X because often in conversation a person’s identity is known by the participants. Think about it: would any of your friends or family members unironically say speaking as a(n) X? When people are being interviewed, their identity is often spelled out before the interview starts. And of course there are many other ways to indicate aspects of your identity in conversation without saying speaking as a(n) X.

I don’t recommend you read Brooks’ column (read this instead). It’s bad. It’s by a white guy who claims he doesn’t understand identity politics. Come on, David. You have a column in the New York Times. You get it. You just don’t want to. If you really need some help, give up your column and hand it to a Latina person, or a queer person, or a Jewish person. Then sit back and read what they have to say. I guarantee it won’t “Speaking as a(n) X…”



James Gleick has a recent article in the New York Times about Autocorrect (“Auto Crrect Ths!” – Aug. 4, 2012), that bane of impatient texters and Tweeters everywhere. Besides recounting some of the more hilarious and embarrassing autocorrections made, he very poignantly tells how Autocorrect works and how it is advancing as computers get better at making predictions.

But in the second to last paragraph, he missteps. He writes:

One more thing to worry about: the better Autocorrect gets, the more we will come to rely on it. It’s happening already. People who yesterday unlearned arithmetic will soon forget how to spell. One by one we are outsourcing our mental functions to the global prosthetic brain.

I don’t know whether Mr. Gleick’s writing was the victim of an editor trying to save space, but that seems unlikely since there’s room on the internet for a bit of qualification, which is what could save these statements from being common cases of declinism. Let me explain.

“People who yesterday unlearned arithmetic” probably refers to the use of calculators. But I would hesitate to say that the power and ubiquity of modern calculators has caused people to unlearn arithmetic. Let’s take a simple equation such as 4 x 4. Anyone conducting such an equation on a calculator knows the arithmetic behind it. If they put it in and the answer comes back as 0 or 8 or 1 or even 20, they are more than likely to realize something went wrong, namely they pressed the minus or plus button instead of the multiplication button. Likewise they know the arithmetic behind 231 x 47.06.

Mr. Gleick writes implies that the efficiency of calculators has caused people to rely too much on them. But this is backwards. The more difficult that calculations get, the more arithmetical knowledge a user is likely to have. Relying on a machine to tell me the square root of 144 doesn’t necessarily mean I “unlearned” arithmetic. It only means that I trust the calculator to give me the correct answer to the equation I gave it. If I trust that I pressed the buttons in the right order, the answer I am given will be sufficient for me, even if I do not know how to work out the equation on pen and paper. I doubt any mathematicians out there are worried about “unlearning” arithmetic because of the power of their calculators. Rather, they’re probably more worried about how to enter the equations correctly. And just like I know 8 is not the answer to 4 x 4, they probably know x = 45 is not the answer to x2 + 2x – 4 = 0.

Taking the analogy to language, we see the same thing. Not being able to spell quixotic, but knowing that chaotic is not the word I’m looking for, does not mean that I have lost the ability to spell. It merely means that I have enough trust in my Autocorrect to suggest the correct word I’m looking for. If it throws something else at me, I’ll consult a dictionary.

If the Autocorrect cannot give me the correct word I’m looking for because it is a recent coinage, there may not be a standard spelling yet, in which case I am able to disregard any suggestions. I’ll spell the word as I want and trust the reader to understand it. Ya dig?

None of the infamous stories of Autocorrect turning normal language into gibberish involve someone who didn’t know how to spell. None of them end with someone pleading for the correct spelling of whatever word Autocorrect mangled. As Autocorrect gets better, people will just learn to trust its suggestions more with words that are difficult to spell. This doesn’t mean we have lost the ability to spell. Spelling in English is a tour de force in memorization because the spelling of English words is a notorious mess. If all I can remember is that the word I’m looking for has a q and an x in it, does it really mean I have unlearned how to spell or that I have just forgotten the exact spelling of quixotic and am willing to trust Autocorrect’s suggestion?

Learning arithmetic is learning a system. Once you know how 2 x 2 works, you can multiply any numbers. The English spelling system is nowhere near a system like arithmetic, so the analogy Mr. Gleick used doesn’t really work for this reason either. But there is one thing that spelling and arithmetic have in common when it comes to computers. Calculators and Autocorrect are only beneficial to those who already have at least a basic understanding of arithmetic and spelling. The advance of Autocorrect will have the same effect on people’s ability to spell as the advance of calculators did on people’s ability to do arithmetic, which was not really any at all.

By the way, I once looked up took (meaning the past tense of take) in a dictionary because after writing it I was sure that wasn’t the way to spell it. And that’s my memory getting worse, not my Autocorrect unlearning me.

[Update – Aug. 6, 2012] If our spelling really does go down the drain, it should at least make this kind of spelling bee more interesting (if only it were true).

Schott’s Vocab I Hardly Knew Ye

To die, to sleep;
To sleep: perchance to dream: ay, there’s the rub;

I didn’t want to believe it, but it seems it’s true. Schott’s Vocab of the New York Times is no more. In the final post, entitled Such Sweet Sorrow, Ben Schott writes:

After two and a half years, thousands of posts and tens of thousands of comments, Schott’s Vocab is closing its doors.

I first came across the wonderful Schott’s Vocab blog in January of this year. I found the famous Weekend Competition, which asked for New Year’s linguistic resolutions. I promptly told them I had resolved to speak more like Snoop Dogg. To my surprise and amazement, my comment was highlighted. I was hooked.

Truth be told, I would have been hooked anyways. Highlighting my comment was merely like adding icing to my cake. Schott’s Vocab was a daily source of mirth for language nuts. And I loved it.

Fo’ shizzle my blizzle.

All the Fake News That’s Fit to Reprint

In yet another case of a news organization being duped by The Onion, the New York Times printed a picture of a fake Tiger Beat magazine with President Barack Obama on the cover, even though the picture was straight out of The Onion and therefore totes fake.

As Joe Pompeo notes in a Yahoo! News blog, this is certainly not the first time this has happened:

Back in 2002, the Beijing Evening News fell for a gag about the U.S. Congress demanding a new Capitol building with a retractable dome. In 2009, two Bangladeshi newspapers were duped by The Onion’s spoof validating fringe conspiracy theories that the 1969 moon landing was a hoax. More recently, Fox Nation fell for the outlandish headline, “Frustrated Obama Sends Nation Rambling 75,000-Word E-Mail.”

But still, at least one question remains: What in the hell was the New York Times doing printing a picture of Tiger Beat Magazine? According to their redaction, the article it appeared in was “about how the original teen-girl tabloid has remained virtually unchanged since its inception in 1965.” So? Is this what waits for me behind that New York Times paywall? Fucking articles about Tiger Beat magazine – articles about Tiger Beat Magazine that aren’t even fact-checked. Christalmighty.

In related news, I’m currently working my way through The Onion’s Our Front Pages book. I highly recommend it.