Press Box

Correct Me If I’m Wrong

Errors and the culture of correction in American newspapers.

In unspooling millions of square yards of newsprint to correct the errors and document the fabrications and plagiarisms of its former reporter, Jayson Blair, the New York Times may have set some sort of dubious record for the grandest correction in the history of newspapers. The investigation continues: On June 11, the newsprint edition of the Times pointed readers to its Web site where the editors ate another 1,200 words of Blair crow about stories he got wrong, made up, or stole.

The Timesreflex to collect and publicize the evidence of Blair’s fraud is laudable, as are the plans afoot at Times and other newspapers to shore up accuracy and apprehend future saboteurs. But before any new accuracy committees are empanelled inside newspapers or the Poynter Institute convenes a summer seminar on errors and fraud, we should ask the question: How error-ridden are American newspapers?

By one measurement, the New York Times is the most inaccurate newspaper in America, running seven or eight corrections per day. In 2002, it ran a total of 2,867 corrections. Compare this count to that of the Washington Post, which publishes a similar number of words each day but logged only 1,006 corrections last year. Or see the Chicago Tribune, whichrecorded 678 corrections in 2002.

But is the Times more than twice as error prone as the Post? Four times sloppier than the Trib? The raw correction count tells only part of the story. Since 2000, the Times has provided instructions in the daily correction box on how to request a correction via e-mail or toll-free phone. Post and Trib readers must guess where to send requests. Either by accident or design, the Post suppresses its official error count with its weekly “Free for All” page, where readers sound off about Post goofs and alleged miscues. Many of these errors go uncorrected in the error box, Post ombudsman Michael Getler writes in this excellent April 27 column about corrections.

Daily Times circulation (1.1 million) towers over that of the Post (746,000) and the Trib (679,000), which means many more eyes scrutinize the Times for errors on any given day. The number doesn’t factor in the Times’ immense Web readership (9.5 million visitors in March, the largest of any newspaper in America) or the millions who read Times stories in their hometown newspaper, here and abroad, via the New York Times News Service. Many of those readers blow the whistle on errors. Demographics help make the Times the corrections lead, too. Times readers are better educated (67 percent are college grad/post grad) than Trib readers (36 percent), making them more reliable spotters of goofs. And because the Times tackles more challenging material than most newspapers, it may stumble more frequently. Put all these factors together, and it’s no wonder the Times correction box swells.

(Cutting against my thesis is the relatively low number of corrections published by the Wall Street Journal. The Journal’s large circulation and bright readership should predict as many or more corrections than the Times, but this week the paper filed 14 corrections, or about three per day. The Journal’s editor likes to say his paper puts a high premium on accuracy because readers trade stocks based on what they read there. I’ll be visiting the Journal’s corrections policy another time.)

When you place the number of Times corrections into context of the number of words printed each year—around 60 million—you don’t ask “Why so many?” but rather “Why so few?” The paper averages one correction every 21,000 words (rule of thumb: 21,000 words equals five long magazine features). It’s a fair bet that additional hundreds or even thousands of mistakes go unnoticed and uncorrected every year. Precisely how many corrections elude the Times is anybody’s guess, but the paper routinely corrects errors committed years and sometimes decades ago. For instance, in 1969, the paper corrected an error made in a 1920 piece about rocket inventor Robert H. Goddard’s theories. The article ridiculed Goddard’s view that a rocket could travel in the vacuum of space.

The Times may owe its status as the corrections king to A.M. Rosenthal, who as the paper’s top editor launched an aggressive campaign in 1972 to detect, right, and highlight Times errors. The Times corrections culture nurtured by Rosenthal seems to revel in correcting every misspelling, transposed digit, historical inaccuracy, and boner. The Times so loves its fallibility that it authorized a collection of its wildest and most shameful errors in the book, Kill Duck Before Serving. Here are a few of my favorite Times corrections from Kill Duck Before Serving. (Click here for more.)

April 25, 1981
An article about decorative cooking incorrectly described a presentation of Muscovy duck by Michel Fitoussi, a New York chef. In preparing it, Mr. Fitoussi uses a duck that has been killed.May 30, 1993
Because of a transmission error, an interview in the Egos & Ids column on May 16 with Mary Matalin, the former deputy manager of the Bush campaign who is a co-host of a new talk show on CNBC, quoted her incorrectly on the talk show host Rush Limbaugh. She said he was “sui generis,” not “sweet, generous.”April 7, 1995
Because of a transcription error, an article about Senator Alfonse M. D’Amato’s remarks about Judge Lance A. Ito misquoted the Senator at one point. In his conversation with the radio host Don Imus, he said: “I mean, this is a disgrace. Judge Ito will be well-known.” He did not say, “Judge Ito with the wet nose.”Jan. 10, 2001
An article about a new sports drink made from a synthetic form of the juice that helps giant Japanese hornets fly faster and farther misstated the amount of weight the hornets can lift. It is three grams, not kilograms.

Who infects the Times with errors? Reporters, mostly. At the Times and most newspapers, reporters act as their own fact-checkers and the paper’s 160-person copy desk—”probably the largest in American journalism,” writes Times Assistant Managing Editor Allan M. Siegal in the introduction—filters whatever bloopers it can as the torrent of copy moves from keyboards to the presses. According to Siegal, one-third of the 138,000 words in the average New York Times are written between 4 p.m. and 9 p.m., after which the presses start printing the paper’s early editions. “In the ‘hard-news’ departments of the paper at that hour, 65 copy editors hold the fort. Ask those copy editors about detailed fact-checking, and they will tell you about trying to drink from a fire hose or bail Lake Michigan with a teaspoon,” Siegal writes.

Old-school journalists believe that newspaper accuracy went to hell when computer technology replaced the typesetter/compositor system in which copy typewritten by reporters was turned into printable type by craftsmen. One typesetter complained to London’s Guardianlast year that in the pre-computer age, typesetters used to catch trivial errors that the copy desk missed. “I would like a tenner for every time I corrected Haversham to Havisham or avoided a split infinitive,” he said. Having worked under both systems, I’d never go back to the old way because even the best typesetter can introduce more errors than he deletes.

How could the Times be made more accurate? One way would be to force Times reporters to memorize Chapter 18 of Kill DuckBefore Serving, which lists frequently misspelled names: Jackson Pollock (at least 66 times in the Times); Dan Aykroyd (at least 45 times); Georgia O’Keeffe (at least 68 times); Aaron Copland (at least 27 times); Philip Morris (at least 128 times); Frederick Law Olmsted (at least 53 times); and Edgar Allan Poe (at least 82 times).

Should the Times and other newspapers employ extra platoons of deadline copy devils to banish 10 percent or 15 percent of its trivial errors each year, or should readers learn to live with the error rate the way we live with traces of arsenic in well-water? (Less than 0.01 milligrams per liter is the current WHO standard.)

Readers should recognize that in journalism as in life, some number of mistakes are unavoidable. And some of those mistakes, while deplorable, don’t matter all that much. Because language is a beautifully redundant communications medium, little or no information is lost if a publication erringly refers to the “University of Indiana” or a “styrofoam” coffee cup.

There are many things newspaper editors can do to raise standards of accuracy, but outside of the serial sabotage by scalawags like Jayson Blair and Christopher Newton, there’s no “accuracy crisis” for newspapers to address. Recognizing the inevitability of error, the most important thing newspapers can do is remain accountable for what they get wrong and make it painless for readers to request corrections. I’m talking to you, Washington Post.


Surely one offends the accuracy gods by writing about corrections. If I’ve made a goof, make sure to send me an e-mail ( and please cc our corrections czar as [June 21 addendum: Sure enough, I misspelled Allan M. Siegal’s last name three times in one paragraph! The corrections have been made, and I am spending the rest of the weekend in the doghouse.]