When is an apology not enough? What is the difference between a clarification and a correction or an admission of error? Is a retraction the same as a recantation? To recant is to withdraw or disavow a declared belief, as in renouncing a philosophy or abjuring fealty to a religion. The most famous recantation in history was that made by the Italian astronomer Galileo Galilei, when threatened by the Inquisition in 1633 with dire punishment for advocating the theory of Copernicus that the earth was not the center of the universe but that it moved around the sun: "I abjure, curse and detest the aforesaid errors and heresies," swore the intimidated scientist, legendarily adding under his breath, "eppur si muove" -- "but it moves." (That retraction of his recantation is apocryphal, probably the starry-eyed wish of subsequent astronomers.)
To retract, from the Latin for "to draw back," is directed to a specific statement more than a body of work. The most famous retraction this year was made by Newsweek magazine after it apologized for a portion of an article alleging that an internal military investigation had uncovered an instance of desecration of the Koran. The article was seized upon by an anti-American Pakistani to trigger demonstrations that cost 17 lives.
At first, the magazine issued an apology -- an expression of regret -- pointing out that its single anonymous source no longer recalled if the allegation was in a certain report. The White House press secretary, Scott McClellan, said that it was "puzzling" that Newsweek had not retracted the article. Next day, Mark Whitaker, the magazine's managing editor, told the New York Times that "it seemed that people felt like we weren't apologizing. In order for people to understand we had made an error, we had to say `retraction' because that's the word they were looking for."
The embattled editor was being precise: Retraction is a weighty word, in this case asserting that the report was mistaken and forthrightly attempting to mitigate the great damage done. The New York Times noted that "Mr. Whitaker contrasted his action with that of CBS News when it refused to back down immediately last year from a report that raised questions about President Bush's National Guard service." At that time, Dan Rather said about documents that many bloggers were the first to accuse of being forgeries that "we can no longer vouch for their authenticity." He said, "Personally and directly I'm sorry," but neither the newscaster nor the network used the word retraction.
That's because the current meaning of retract goes well beyond "we're not sure of this, so we're withdrawing it with our apologies," which is a mild form of reputation reparation; instead, retract means "we made a mistake and we take back what we published or broadcast so you can see we weren't being malicious."
Though libel law differs from state to state, a prominently displayed retraction can be evidence of a lack of "actual malice," one of the standards established in the Supreme Court's famous 1964 New York Times v. Sullivan decision. The Yale law professor Jed Rubenfeld informs me that "a retraction generally admits publication of inaccurate (ie, false) information but does not admit malice." Indeed, a publisher's failure to make a retraction may help a public figure to establish such malice needed to win a defamation case. (Belatedly for a former polemicist, I've been reading Sack on Defamation, by Appellate Justice Robert D. Sack, where I found the Galileo footnote.)