Approximate quotations
« previous post | next post »
I need to apologize for causing some confusion. My recent posts on journalistic quotation practices ("Jonah Lehrer, Bob Dylan, and journalistic unquotations", 8/3/2012; "More unquotations from the New Yorker", 8/4/2012) dealt with two issues at once: journalistic carelessness and journalistic deceit. And some readers seem to have concluded that I meant to treat all carelessness as a form of deceit.
Journalists are indeed routinely careless about quotations, and that tolerance for carelessness makes it harder to regulate deceit, since accusations of journalistic deceit often come in the form of accusations about fabricating quotations. But most cases of "approximate quotations" are innocent enough, except for their role in perpetuating this culture of carelessness.
I should add that quotations can also be misleading or false without being maliciously deceitful, when the writer misunderstands things or wants to simplify a complex issue. This is especially common in writing about science and engineering, where it's difficult to separate deceit from carelessness, confusion, or the simple desire to tell a good story.
Several working journalists objected strongly to the way I characterized their profession's attitude towards direct quotes. Some of them just objected and left it at that:
I wrote for magazines for many years, and I did not make up quotes. Ever. If I had, and had been found out, I would have been dropped by every editor I worked with.
Others agreed with my picture of the situation while offering some explanations. Thus Sid Smith:
Transcribing tapes is very time-consuming, so it's common for journos to tape an interview and yet rely on their contemporaneous notes, or even just their memories, to write the quotes. This can be very wrong or just a little bit wrong – but it's wrong.
And others simultaneously agreed and objected. Thus the always-interesting Kyrie O'Connor, "Zen and the art of quoting people", Houston Chronicle 8/7/2012:
What astonishes me but maybe shouldn’t is that this would be an issue at all. I suppose Jared Diamond is not a journalist (a work I now cannot say without putting the effete Emily-Mortimer-in-”The-Newsroom” spin on it: JAHnalist). But because I am, I just sort of assume, because I had it pounded into me back in the day when yelling at young reporters came as naturally to editors as flicking their cigarette ashes into the keyboard, that quotes are quotes, and one’s job is to get them down as accurately and naturalistically as possible.
Seriously. I have only recently had someone write to me asking if I had actually talked to a person I quoted. Answer: Yes. Always yes unless otherwise noted.
Now, I rarely record interviews unless the story is going to be in a q&a format or some other that would require paragraph-length quotations. I suppose that makes me old-fashioned. I also supposed it means I get the odd word wrong now and again, for which I apologize. I think I can speak for my colleagues past and present who do the same.
This summarizes the confusion that I started this post by apologizing for. If you write quotes based on your memory or your notes, it's pretty much guaranteed that the quotes will be approximate ones. And most journalists continue to operate that way, although appropriate use of modern technology now makes recording and checking fast and easy.
We can pick a random event — say, Mitt Romney's stump speech in Des Moines earlier this week — and check my claim that most journalistic quotes are approximate.
Note that this is the sort of case where approximate quotations are least likely to happen. Romney's speech was recorded and posted on CSPAN later the same day. Dozens of journalists produced dozens of stories about the speech, so the same bits are likely to be quoted more than once. Various political partisans are sure to monitor the stories and to object to any perceived inaccuracies. And the journalists covering the event know all of this. The level of scrutiny and accountability in this case is sure to be much higher than it would be with respect to the quotes in an individual interview with some random source, where there's no independent evidence about what was said, and no one really cares except the source and her mother.
Despite this, all of the quotes (that I checked) in media coverage of Romney's Des Moines were approximate. Here's one passage that was widely quoted:
"In a very careful executive action, he removed the requirement of work from welfare," the GOP presidential candidate said at an early-morning rally at a downtown Des Moines school.
"It is wrong to make any change that would make America more of a nation of government dependency. We must restore, and I will restore, work for welfare!" he said to loud cheers and applause from the crowd of about 700. (USA Today)
“With a very careful executive action, [Obama] removed the requirement of work from welfare,” Romney said. “It is wrong to make any change that would make America more of a nation of government dependency. We must restore it, and I will restore work in welfare.” (Washington Post)
"It is wrong to make any change that would make America more of a nation of government dependency," Romney said. "We must restore work in welfare." (ABC News)
On Wednesday, Romney doubled-down on the attack, telling an audience in Des Moines, Iowa that Obama’s new rules will help create a “nation of government dependency.”
“I will restore work to welfare,” Romney added. (NY Daily News)
What he actually said was different from all of these:
Now he's president.
And just a few days ago, he uh he put that original intent in place.
With a very careful executive action,
he removed the requirement of wel- work from welfare.
It is wrong
to ch- to make any change that would make America
more of a nation of government dependency.
We must restore, and I will restore, work into welfare.
It makes sense to remove the filled pause and the false starts, and introduce standard paragraphing, giving us something like this:
Now he's president. And just a few days ago, he put that original intent in place. With a very careful executive action, he removed the requirement of work from welfare.
It is wrong to to make any change that would make America more of a nation of government dependency. We must restore, and I will restore, work into welfare.
But still, every one of the journalistic quotations was approximate.
Let's look at a few more of the quotations in that same Washington Post story:
“When it comes to the spirit of America, I want to restore the spirit of independence. I do not want to install a spirit of dependence on government, and that’s the direction we’re going,” Romney told several hundred supporters in a steamy high school auditorium in Des Moines.
What he actually said:
But when it comes to the spirit of America,
I want to restore the spirit of independence.
I do not want to install a spirit of dependence on government.
And that's the direction we're headed in.
Again an "approximate quote". Is "the direction we're going" really very different from "the direction we're headed in"? No. But it's still a paraphrase, not a quote.
Another approximate quotation, again from the same WaPo article:
“When that kid gets the honor roll, I realize to get to school they’ve got to go on a bus. And the bus driver is driving the bus."
What Romney actually said:
When that kid gets the- gets the honor roll
I realize to get to school he had to go on a bus
and- and the bus driver had to drive the bus
Or, cleaning up the disfluencies:
When that kid gets the honor roll, I realize to get to school he had to go on a bus. And the bus driver had to drive the bus.
Web logs are no better. Here's part of the coverage at Caffeinated Thoughts ("Stimulating Christian Conservative News and Commentary"):
Closing his speech he said, “I want to restore a spirit of independence, I don’t want to install the spirit of dependence on government… (Obama) looks at government as the source of our greatness, I see our people as being the source of our greatness… I don’t want to change America I want to restore us to our principles.”
What this corresponds to in the speech:
I want to restore the spirit of independence.
I do not want to install a spirit of dependence on government.
[466 words omitted]
I- I- you know he- he looks at government as the source of our greatness.
I look at the American people as the source of our greatness.
[152 words omitted]
I don't want to change America.
I don't want to turn America into something we wouldn't recognize.
I want to restore to America the principles that made us the hope of the earth
And local papers don't seem to be much different from the national ones. Here's the Waukee Patch:
"I want to make sure that our adults and kids have the skills to succeed," Romney said. "What happens here at Central Campus is that young people are able to come here and get skills that let them get the jobs they have today."
What he actually said:
I want to make sure that our adults and our kids
have the skills to succeed.
What happens here at Central Campus is that young people
and some a little older
are able to come here and get skills of a wide variety
that will allow them to be successful in the jobs of today.
Again, the same journalist's version of another quote:
Romney's final point centered around the idea that small businesses must be championed and businesses need less regulation.
"My priority is more jobs and more take home pay, so I want to help do that. So how do I do that? I want to make sure the tax rates aren't raised, and the president would raise that."
What he actually said:
My priority is more jobs and more take-home pay
for middle income Americans so
I want to help small business. How do I do that?
Well, I want to make sure that the tax rates on small business aren't raised.
The president would raise them.
I hope that this establishes my point, which is that direct quotes in newspaper and magazine articles are almost always approximate at best.
Are the approximations malicious? Usually not — they usually arise simply because the quotes are based on writers' notes and memories. Do the approximations matter? Usually not. But then there's a long slide through confusion and bias into intentionally misleading quote-mangling and outright fabrication.
And modern technology makes it easy to record most speeches and interviews, and to remove most forms of approximation from quotations. Isn't it time for this aspect of journalism to limp into the 21st century?
Some other relevant LLOG posts:
"What did Rasheed say?", 6/23/2005;
"Ipsissima vox Rasheedi", 6/24/2005; "
Ritual questions, ritual answers", 6/25/2005;
"Down with journalists!", 6/27/2005;
"Bringing journalism into the 21st century", 6/30/2005;
"More comments on quotes", 7/1/2005;
"Linguists beware", 7/9/2005;
"Quotes from journalistic sources: unsafe at any speed", 7/9/2005;
"'Quotations' with a word error rate of 40-60% and more", 7/30/2005;
"This time it matters", 8/13/2005;
"'Approximate' quotations can undermine readers' trust in the Times", 8/27/2005;
"Imaginary debates and stereotypical roles", 5/3/2006;
"Journalists' quotations: unsafe in any mood", 5/24/2007;
"In president, out president, fake president", 12/5/2008;
"Audio photoshopping at NPR", 5/31/2007;
"Filled pauses and faked audio", 12/28/2008;
"More (dis)fluency and (in)coherence", 12/31/2008;
"Egregious fabrication of quotes at the Sunday Times?", 1/29/2010.
Sid Smith said,
August 11, 2012 @ 7:23 am
"Others agreed with my picture of the situation while offering some excuses. Thus Sid Smith: 'Transcribing tapes is very time-consuming, so it's common for journos to tape an interview and yet rely on their contemporaneous notes, or even just their memories, to write the quotes. This can be very wrong or just a little bit wrong – but it's wrong.' "
Yikes. I offered excuses? You didn't get my quote wrong (pretty difficult in this format, of course), but your interpretation is as odd as any of the examples you are denigrating.
[(myl) Sorry. In describing your comment, I should have said, "offered explanations". Both you Kyrie O'Connor (and similarly Linda Seebach in this old post) explain that working with recordings (or "tapes", as you call them) is a time-consuming process, and that reporters facing a deadline don't have the time to deal with it. You go on to emphasize that the resulting approximations are wrong, which Kyrie and Linda didn't do.
Anyhow, I changed "excuses" to "explanations" in the body of the post. Does that answer the objection?]
David Faulder said,
August 11, 2012 @ 7:50 am
Perhaps we need ~paraphrase marks~ to add authority to something that does not quite deserve it, and thus leave quotation marks for genuine quotation?
Sid Smith said,
August 11, 2012 @ 8:03 am
Yep. Quite happy. ;-)
"Both you Kyrie O'Connor (and similarly Linda Seebach in this old post) explain that working with "tapes" is time-consuming, and that reporters facing a deadline don't have the time to deal with it."
I didn't, as it happens, say that journos avoid transcribing because of deadlines: unfortunately, as you have suggested, it's often just part of the culture. Your compilation of Romney "quotes" is vivid and shaming.
[(myl) Oops, sorry again. Jeez, this journalism stuff is hard :-). You wrote:
Transcribing tapes is very time-consuming, so it's common for journos to tape an interview and yet rely on their contemporaneous notes, or even just their memories, to write the quotes.
I interpolated the idea that the working under deadline offers a reason not to take the extra time — but as you say, the practices don't seem to be different when the time pressure is absent.]
Sid Smith said,
August 11, 2012 @ 8:36 am
No prob. ;-)
I didn't stress enough how good that Romney stuff is. Anyone running a journalism course should plagarise it at once.
Sid Smith said,
August 11, 2012 @ 8:37 am
Blimey. I've gone all red.
[(myl) Typo in my threaded-comment approximation technique. Fixed now, I hope. I wish we had proper threaded comments — maybe after I update our antique WordPress installation…]
Eric P Smith said,
August 11, 2012 @ 10:11 am
Did Sid Smith offer excuses?
Interesting noun, 'excuse'. Oxford Dictionary Online says, "a reason or explanation given to justify a fault or offence". Such a reason or explanation given may be adequate or inadequate, whence I suppose the two different senses of 'excuse'. When I was at school, one teacher used to parry explanations for lateness, poor work etc as "That's no excuse" while another teacher used to say "That's just an excuse". I felt like saying to them, "Make up your minds! Is it an excuse or is it not?"
I guess myl wrote 'excuse' meaning 'adequate excuse' and Sid read it as meaning 'inadequate excuse'.
Nick Lamb said,
August 11, 2012 @ 12:25 pm
An example of the importance of quoting:
"a work I now cannot say"
That's clearly a typo, the writer intended "word" not "work" here. But who made the mistake?
Ideally we would have transclusions, but the web doesn't natively support them (blame Sir Tim) so we have to resort to the next best thing, a link. We can consult the link and see (at least, at the time of writing this comment) that the mistake is present in the original and has been faithfully copied here by Prof Liberman.
Clayton Burns said,
August 11, 2012 @ 12:39 pm
The opportunity costs of recording and transcribing as a habit should be examined.
A linguistically competent human should be able to accept the challenge of listening minutely and getting it down.
There is reductionism in relying on transcribing.
I like the old-fashioned reporters who want to do it the natural way.
A good sign with Obama on the vp slip is that he picked it up immediately. His linguistic working memory worked. The other man, consistent with his rather painful handling of the English language, could not get his working memory to work. So he left the impression that he was handing off his candidacy. Perhaps an unconscious slip. Why his staff did not anticipate this problem is a mystery. Language Log is indicated.
Reporters should be able to record and transcribe effectively if needed. But they should also have a powerful skill in getting it down without the aid of technology.
Bill Marvel, who helped Burgin with his Marine account of the war in the Pacific, e-mailed me that the recording and transcribing process was harrowing.
Sid Smith said,
August 11, 2012 @ 1:46 pm
"A linguistically competent human should be able to accept the challenge of listening minutely and getting it down"
Not sure what you mean, unless it's that reporters should have lightning shorthand. Well, possibly: but I reckoned that a tape was far better for both accuracy and for maintaining the flow of the interview.
Mark F. said,
August 11, 2012 @ 2:34 pm
Does the ability to record interviews make transcribing them fast enough to justify doing it as a routine practice? Finding a quote in a sound file isn't easy at all. Time spent doing that is time that could be spent checking other facts, digesting information, or making articles clearer.
In response to what I just wrote, it occurs to me that radio reporters are forced to spend their time in just that way.
Sid Smith said,
August 11, 2012 @ 3:16 pm
@ Mark F.
"Does the ability to record interviews make transcribing them fast enough to justify doing it as a routine practice?"
Yes, yes and yes. What could be a bigger priority than getting the quotes right? It's a moral obligation to both reader and interviewee.
The misquotes that mjl has assembled are a disgrace – shoddy, sloppy, lazy.
[(myl) Many programs and apps, for laptops or PDAs or tablets, allow you to add bookmarks or labels while recording or listening. This means that someone recording an interview or a speech can mark points of interest to make it easier to find things later on. Once you have a recording, many programs and apps allow you to increase playback speed by up to a factor of 2 or 3, which also makes locating relevant regions faster.
I don't know of any decent integrated systems (at least none suitable for the uses we're talking about) for using speech recognition to segment and search for content. But such things will be available before long.
But even without any of this fancy stuff, it's not that hard to find and check the bits of a speech or interview that you want to quote. Let's take the case of Mitt Romney's little "next president" mistake this morning. I captured 40 minutes of the audio on the my laptop (from 9:00 to 9:40) from a radio station's internet stream, while I was busy doing other things. I sometimes listened in the background but mostly didn't. Some time later, I saw that Ben Zimmer put up a post about the error, and decided to check the quotation, since it didn't agree with the portion given in the vido clip in Knox's post. So I sat down to find the crucial passage in that 40-minute long recording. I could guess that the overall pattern would be (1) Whoever (2) Romney (3) Ryan, and that the critical point would come at the end of Romney's segment. Using audacity (the program, not the attitude) and jumping around in the file, I was able to find the relevant passage in less than a minute of searching. Checking the exact quote took maybe another minute. Composing my comment took longer than both of these steps put together.]
Clayton Burns said,
August 11, 2012 @ 3:41 pm
http://www.blinkx.com/videos/mark+liberman
This tool could be of some use for political speech.
Searching audio files should not in theory be difficult.
I am willing to concede that recording and transcribing can be valuable for fast-moving speech, but I think that there is some exaggeration about how hard it is to listen and write down what people say.
I can do supreme court-style discourse, but it takes concentration.
What we want is reporters and editors with sharp linguistic skills.
Those who cannot capture natural speech with a notebook need to be retrained, I think.
[(myl) Excellent linguistic memory is a fine thing, but the idea that quotes shouldn't be checked against the original source is silly. If you're quoting a passage from a famous historical text, would you refuse to check the original on the grounds that "what we want is reporters and editors with sharp linguistic skills"? I don't think so, especially now that it's usually so easy to look such things up. In my opinion, the same principles apply to quotations from audio texts.]
Sid Smith said,
August 11, 2012 @ 4:21 pm
"Once you have a recording, many programs and apps allow you to increase playback speed by up to a factor of 2 or 3, which also makes locating relevant regions faster"
I still have the cassette 'Walkman"-sized recorder I used back when I was reporter rather than a sub-editor. It must be 15 years old, but playback speed can be adjusted both up and (very useful during transcription) down. I used to contemplate jotting down the number on the tape counter for any interesting bits, but (as I mentioned above) just keeping the flow always seemed more useful: it's handy, after all, if the interviewee forgets it's an interview.
Clayton Burns said,
August 11, 2012 @ 5:06 pm
–I don't know of any decent integrated systems (at least none suitable for the uses we're talking about) for using speech recognition to segment and search for content. But such things will be available before long.
More information would be interesting.
What I have found is that if I write down what is being said it helps with concentration and with seeing how various bits of information fit together.
I have no objection to recording and checking quotes.
The way information develops, it is a good idea to be working through the content while keeping a written record. That way, if something seems out of focus, you can often ask a friend about it, or interview the person who said it.
It would seem silly to me to go back to the office, play the recording, realize you do not grasp the implications of something, then try to retrace the steps.
What I am saying is based on my experience. Being able to rely on notes and recordings fits the accounts of attention, working memory, language, and executive function in "Philosophy, Neuroscience and Consciousness," by Rex Welshon (chapter 7).
tpr said,
August 13, 2012 @ 4:15 am
Mark, the embedded audio widget for the "When that kid gets the- gets the honor roll" is linked to the wrong audio clip.
[(myl) Thanks. Fixed now.]
Martyn Cornell said,
August 13, 2012 @ 8:00 am
Clayton Burns: "I think that there is some exaggeration about how hard it is to listen and write down what people say."
Having done this – taking down verbatim peoples' speeches at meetings, in courtrooms and so on – for the past 38 years, I can tell you that unless you have 150wpm+ shorthand (which I don't), it is very easy to accidentally find yourself paraphrasing a speaker's remarks, using phraseology of your own rather than the speakers', as you try to keep up.with what is being said. Comparing some of those transcripts above with some of the reporters' versions, I recognise exactly that problem: the reporter simply wasn't fast enough, and wrote down their impression of what had just been said, using their own terminology.
Personally I try now to record everything in an interview and then transcribe it all: yes, it's laborious to get the transcription done, but I know everything quoted is the speaker's exact words, not my impression of their words.
Will Keats-Osborn said,
August 13, 2012 @ 6:37 pm
Interesting and relevant anecdote: A self-described "human tape recorder" with "a talent for mentally recording lengthy conversations," Truman Capote thoughtfully included in his book Portraits and Observations an opportunity to verify this claim, as two separate depictions of an encounter with Willa Cather were included in the book, one from 1979 and one from 1984.
The first one:
And she wanted to know what American writers I liked. "Hawthorne, Henry James, Emily Dickinson…" "No, living." Ah, well, hmm, let's see: how difficult, the rival factor being what it is, for a contemporary author, or would-be author, to confess admiration for another. At last I said, "Not Hemingway—a really dishonest man, the closet-everything. Not Thomas Wolfe—all that purple upchuck; of course, he isn't living. Faulkner, sometimes: Light in August. Fitzgerald, sometimes: Diamond as Big as the Ritz, Tender Is the Night. I really like Willa Cather. Have you read My Mortal Enemy?"
With no particular expression, she said, "Actually, I wrote it."
The second one:
Whereupon I told her all about myself. My age. The fact I was born in New Orleans, and that I was an aspiring writer.
Really? What writers did I admire? (Obviously she was not a New Yorker: she had a Western accent.)
"Flaubert. Turgenev. Proust. Charles Dickens. E. M. Forster. Conan Doyle. Maupassant—"
She laughed. "Well. You certainly are varied. Except. Aren't there any American writers you care for?"
"Like who?"
She didn't hesitate. "Sarah Orne Jewett. Edith Warton—"
"Miss Jewett wrote one good book: The Country of the Pointed Firs. And Edith Wharton wrote one good book: The House of Mirth. But. I like Henry James. Mark Twain. Melville. And I love Willa Cather. My Antonia. Death Comes for the Archbishop. Have you ever read her two marvelous novellas—A Lost Lady and My Mortal Enemy?"
"Yes." She sipped her tea, and put the cup down with a slightly nervous gesture. She seemed to be turning something over in her mind. "I ought to tell you—" She paused; then, in a rushing voice, more or less whispered: "I wrote those books."
Even if I'm charitable enough to concede that maybe he did actually chance upon meeting Willa Cather, his account of this meeting in either case is apparently so fictionalized as to be absurd, and I now have trouble believing that anything Capote ever wrote about happened the way he said it did.
Jonathan D said,
August 13, 2012 @ 7:31 pm
Sid Smith didn't appreciate Mark Liberman adding the interpolation about deadlines in the comments above, but at least it was clear that it was Liberman's summary. It would be much worse if that interpretation had had an effect on an approximate quote!
maidhc said,
August 17, 2012 @ 1:08 am
Today's San Francisco Chronicle has the following direct quotation in a featured article:
"There's a way in which some violinists can slide or use certain kinds of bravado to emulate what an opera singer does."
I'm pretty sure that the interviewee actually said "vibrato", which would make a lot more sense.
[(myl) Publishing the article with that mistake in place is certainly a sign of careless copy-editing. but the mistake itself might be a Cupertino due to typing e.g. "bivrato", or just a malapropism. Either way, it could easily arise in an attempt to carefully transcribe a recording, rather than in approximation from notes or memory. ]
John said,
August 19, 2012 @ 7:59 pm
Any chance reporters here (or elsewhere) are using prepared texts given to them by campaigns, as they do with the State of the Union, for example, and Romney read it wrong?
James Joyce on Writing: “write dangerously” | The Coming of the Toads said,
August 20, 2012 @ 12:23 pm
[…] Menand questions whether what we read in Power's book are the actual words of Joyce or the "gist" of a conversation that took place decades prior to the book's publication. Would Menand have the same complaint if Richard Ellmann, Joyce's highly regarded biographer, was the one recalling the conversations? It seems Menand thinks Powers belongs, if he should be mentioned at all, in a footnote somewhere, his "renown" based on a single book. [An argument ensues, as Gordon Bowker, whose new biography Menand is reviewing, responds, timely, for the question of the journalistic practice of approximating quotes is in the air]. […]
Rob said,
August 21, 2012 @ 12:29 pm
The problem I have as an editor is that often people don't say what they mean. They might use the wrong word, or mess up the grammar, or – increasingly these days – mangle an idiom. Where it's obvious what was meant, and you're not changing that meaning, I don't see the problem in cleaning up the quote somewhat. Otherwise your readers will be distracted by the mangled sentence rather than getting the point of the quote.
Brad Daniels said,
August 21, 2012 @ 8:57 pm
It seems to me that technology has already come up with a solution that is perfect for journalists who like to use written notes: I have a pen (a Livescribe Echo) which works with a special notebook to digitize everything I write, and which can simultaneously record audio. The pen's compelling feature for journalism is that if you make notes while recording, you can later touch the pen to the text you wrote in order to play back the portion of the audio you recorded while making your written notes.
You can of course also upload your notes to a computer to gain access to additional bells and whistles.
Unquotations and How to Avoid Them | Book How-To said,
September 18, 2012 @ 6:43 pm
[…] explores the topic further in follow-up articles here and here. Commenters offer a few possible explanations for inaccuracies—transcribing recorded […]
The Language Log scolds journalists for "approximate quotation" (again) | Schaver.com said,
August 23, 2013 @ 9:13 pm
[…] unquotations", 8/3/2012"More unquotations from the New Yorker", 8/4/2012"Approximate quotations", 8/11/2012"Quote approval and accurate quotation", […]