Department of Corrections


In the January 19 New York Observer, Ron Rosenbaum posed a slew of questions about how print and online media should approach the difficult job of correcting errors that are discovered post-publication, then beseeched media critics to come up with some rough standards. Ron and I are friends, so I’d be remiss to ignore the call.

Rosenbaum’s proposal for increased disclosure goes against traditional newsroom wisdom, which looks askance at self-flagellation and the airing of dirty laundry. But let’s face it, these are the days of dirty laundry, when even The New York Times has become less uptight about its own infallibility and is striving for a more responsive corrections policy.

For example, on January 18, the Times ran an Editors’ Note regarding Ian Buruma’s January 11 review of a book by Shashi Tharoor. Because Tharoor had reviewed Buruma’s novel in The Washington Post in 1991, the recent review violated a policy under which “the Times does not normally assign authors to review each other’s books.” Both reviews were barbed—there is surely more to this story—but the Times deserves credit for the speed with which it’s now admitting its errors. Last Sunday, the Times‘ public editor, Daniel Okrent, even went so far as to correct himself!

The more corrections, the better? Without lowering evidentiary standards, Rosenbaum suggests, journalists can alleviate the shame of error by approaching our work more as a search for truth and less as a crusade to prove we’re always right. In this view, all assertions are provisional and necessarily subject to reshaping as new facts, theories, and agendas emerge. In other words, it’s inevitable that mistakes will be made when we’re gathering all the news that’s fit to print. Rosenbaum suggests that errors are more easily correctable now that most stories, even those published in print editions, also appear online—and more visible, now that bloggers are keeping watch. Ironically, increased self-consciousness about the potential for error increases a publication’s credibility.

To arrive at rough standards, I studied websites maintained by Slate, The Washington Post, and The New York Times. My suggestions: Every publication should have an online corrections page that is easy to find—and easy to navigate. Each correction should be clearly stated and dated, and should carry a link to the online text of the original article. The corrections should be arranged by the date of the articles, in reverse chronological order. In addition to newly published corrections, the page should carry links to the archives. Success means the average reader can find out quickly whether any recent story has been amended and how.

Slate gets the Best Display of Mistakes award. The link to “Corrections” from the home page is a little hard to find, as it rises and falls in the table of contents. But once you get there, corrections are clearly dated and stated, and each one provides a link to the live online text, where erroneous passages are asterisked and corrected at the bottom of the page. There’s a link to each prior batch of errors, as well as an e-mail address where readers can report new ones. (Salon‘s model is also simple and transparent.) scores high in every cate-gory, including comprehensiveness and cross-referencing. In the left column of the home page is a news box in which the heading “Corrections” connects to a page of “Recently Corrected Articles,” with links in reverse chronological order. Each link on that page contains the original headline and date of the corrected article. Click on one of those and you find the live online text with errors left intact. In the right column of the live page, the word “Correction” appears in red in a box, followed by an explanation. One can also find a link to “Corrections From the Post,” which offers an aerial view of links to both corrections and clarifications, those more elaborate rethinkings. The design is intuitive and utilitarian.

By contrast, the corrections page on feels like a Soho boutique—a few boldly displayed items with lots of empty space around them. It’s easy to find the corrections page from the home page, but once you get there, all you find are the fixes published in that day’s paper. The items that appear on the daily corrections page are inconsistently styled, sometimes identifying the exact date and section in which the article appeared and sometimes referring obliquely to an article “yesterday” or “last Sunday.” The text of each correction offers no link to the original live text.

It took me three or four visits to to figure out that they do keep corrections pages from previous days on the site; it’s just that the links are not presented in one place. Finding yesterday’s fixes takes too long to be of much practical use. Likewise, it takes too much diligence to match a Times correction to its corresponding live text. When you do find the live text, corrections have been inserted invisibly, while Editors’ Notes appear at the bottom of the page, even when a story is old enough that one must pay for the full text. That’s good, as is the prominent display of an e-mail address and phone number for the public editor. Now if only the Times would give Daniel Okrent a blog.

Trying to achieve an industry standard of disclosure sounds good in theory, but it conflicts with the goal of charging for online news content. Thus, it’s likely that the Times doesn’t offer a corrections archive because that would require the company to free up online content now available only for a fee. Besides, anyone with a Nexis password can find every correction and Editors’ Note the Times has ever published—but subscriptions are expensive, which is why Rosenbaum calls Nexis a “gated community.” Shouldn’t the library of errors be open to the public, and not just to insiders?

Here’s another thorny issue. It is impossible to establish the truth or falsehood of certain propositions, such as whether the pope said, “It is as it was,” or whether the Dean scream means he is unstable. Perhaps online news sites could address this sort of thing by devoting separate pages to Second- and Third-Hand Assertions, Spin We Fell For, and Hidden Agendas.

Carried to its absurd conclusion, this trend could lead to an All-Corrections website, a universal library of errors. Think of it as one-stop shopping for researchers, a complete inventory of all corrections, all the time. The home page would link daily to every media outlet’s internally generated corrections, with separate pages for Editors’ Notes, ombud blogs, and independent criticism. would depend on elegant design and a crack search engine and would work best as a nonprofit, with a paid webmaster and voluntary media participation. Of course, to maintain credibility, the site would itself have to have an ombudsman, a readers’ forum, and a corrections page . . .

Remember, cyberspace means never having to say you’re sorry.