Let's Get This Straight: How do you retract a story online?

Time, the New Republic and the Cincinnati Enquirer map the high, middle and low roads for dealing with discredited articles on the Web.

Published July 17, 1998 7:00PM (EDT)

When the Cincinnati Enquirer recently decided to disown a much ballyhooed investigative series alleging a variety of misdeeds on the part of Chiquita Brands International, it not only paid Chiquita $10 million and published an abject apology -- it also wiped the stories off its Web site and removed them from its online archive, leaving its apology and retraction to hang forlornly out of context.

Meanwhile, over at Time magazine, editors were also retracting a much trumpeted Time/CNN investigative scoop that claimed the U.S. military had used nerve gas in Laos in 1970. But you can still read this discredited story on Time's Web site -- although not before being directed to Time's abashed retraction and apology.

Back when newspapers and magazines first rushed onto the Web, editors and publishers dreamed of reaching vast new audiences and tapping rich new veins of advertising revenue. In those heady days, how many of them could foresee that their new online presences would also become major legal and ethical headaches for them?

The latest wave of media blunders has put print publications in a quandary that they never faced while they remained in a strictly paper universe. In the old days, if you published a flawed story in one day's edition, you ate crow in the next day's. You couldn't go out and destroy all the already-distributed copies of your publication that contained the offending work, nor did anyone expect you to. The historical record -- including a full detailing of both the publication's original missteps and its efforts to make amends -- remained intact. That meant the public could judge for itself how badly the publication had damaged its credibility, whether the amends it made were forthright or foot-dragging -- and whether they were presented in a spirit of truth-seeking or performed under instructions from craven company lawyers.

The Web raises vexatious new questions about just what constitutes a retraction -- and new temptations for publications that might rather hide black marks on their records than keep dirty linen hanging in their online archives. Of course, Web sites have always faced thorny ethical and technological problems when they need to correct errors. Since a "published" article on the Web is merely a file sitting on the hard drive of a Web server, you can easily go into a Web page and fix a mistake so that there's no record of the original goof. Some sites keep track of every little fix they make, "paper-of-record" style; others just make their corrections with no notice at all. (Here at Salon, our policy has evolved over time: Today, we will simply fix minor errors, such as misspellings of names, soon after publication, without comment; more significant errors of substance are noted prominently on our Letters page.)

But the Time and Cincinnati Enquirer stories weren't simple corrections -- they were high-profile, full-bore retractions of major features. In the Enquirer's case, there are still serious questions outstanding as to whether the Chiquita stories deserved to be disowned -- or the paper simply caved to pressure from a powerful local corporation. Under such circumstances, the alacrity with which the Enquirer scoured its archives of 30 or more Chiquita investigation stories is not only unseemly, it robs the public of the chance to judge the series for itself -- alongside whatever disclaimers, retractions and apologies the newspaper has since deemed necessary.

In any case, the Enquirer can delete its stories from its own site, but it can't keep them off the Net. For one thing, for the moment you can still obtain them (for $1 per story) from directory services such as Northern Light. And individuals on the Net who believe that readers should have the chance to study these stories for themselves have also taken the liberty of reposting them. Once a text is distributed online, it's very difficult to efface it from the Net if someone is passionate about keeping it in circulation -- and willing to take a legal risk.

Given all this, the Enquirer's tactic looks like a model for how not to deal with the Web retraction problem -- and Time's approach, whatever its other missteps in this affair, looks like the way to go. Time has corrected the record by augmenting it rather than erasing it; while it's no longer possible for Time readers to read the original story without being aware of Time's subsequent retraction, it's also not possible for Time to pretend that the goof never happened.

Contrast that forthrightness with the more dodgy approach taken by the New Republic in the Stephen Glass affair. In the wake of the initial revelation that a Glass story for the magazine about teen hackers was a fabrication, curious people online began combing the New Republic Web archive and finding other dubious-sounding yarns in Glass' backlist -- like those on inflatable Monica Lewinsky sex dolls and Alan Greenspan groupies -- that were later shown to be fraudulent, too. At first the magazine left Glass' oeuvre in full view; later it took all the discredited stories down, explaining the move in a letter to readers: "When we post something to our archive, it is being continuously published, and that implies ongoing endorsement of its honesty and truthfulness."

That sounds good, but it gives the magazine a convenient out, a chance to bury the embarrassing incident. There's no reason the magazine couldn't "continuously publish" the stories with an explicit statement that they're not honest and truthful. Time's approach is what the New Republic should have adopted: Leave the stories up, leave the historical record intact and append a note to readers (with appropriate links) explaining the subsequent history. With this method of dealing with errors, journalists can actually make the Web function better than print: After all, you don't find magazines going into the library stacks inserting notes of retraction into bound volumes.

Of course, each time a magazine tries to correct a problem, it opens the door to making new mistakes. If you take a look at the Web page for Time's nerve-gas story, you'll notice something strange: The original story is dated June 15, yet the type directing readers to the retraction says, "Please see the editor's note appearing in the June 13, 1998, issue of Time." Make that "July" -- and no need for an apology.


By Scott Rosenberg

Salon co-founder Scott Rosenberg is director of MediaBugs.org. He is the author of "Say Everything" and Dreaming in Code and blogs at Wordyard.com.

MORE FROM Scott Rosenberg


Related Topics ------------------------------------------