Online news stories, corrections, and keeping up to date
One of the interesting wrinkles of journalism in the online age is the ease of accessing archives. It used to be that back issues were at best scanned to microfiche and stored in some musty, stale vault somewhere. Now, though, anyone with an Internet connection can get to them.
This is not exactly an original observation, but there is an implication to it that doesn't seem to have been discussed very much: News sites can, if they want, issue corrections long after publication. Back when everything was a mass produced hard copy, something that went out wrong was staying wrong forever. A correction could be run in a subsequent edition, but the mistakes that had been printed and shipped were out there for good. There's a certain messiness implied in that, and I think everyone from publishers to readers made allowance for it.
While that dynamic remains for physical editions, the online counterpart of a story is printed in exactly one place - and updates to it are effective immediately. News organizations can (and do) make revisions to electronic versions fairly quickly. From my own observation though, they seem to make corrections using the same model as print: when errors surface fairly quickly.
Once a week or so passes, the chances of a correction being run appear to be very small; after about a month almost zero. On the face of it that's understandable. A news organization could devote all its resources to correcting every last little thing brought to its attention, no matter how far back. Going down that rabbit hole would be crazy.
On the other hand, the ease of retrieving and correcting Web pages makes them something other than the "first rough draft of history." They aren't (or don't have to be) fleeting, indelible first impressions. For issues of great and ongoing importance, a newspaper's site can - and maybe should - reflect changes to the story as it evolves. Subsequent drafts of history can easily be incorporated. Not only would that reflect a decent journalistic ethic of getting things right, it would also be a service to readers who take the time to research these newly liberated archives.
All this began to rattle around my brain as I've been making my way through Dirty Wars by Jeremy Scahill. He extensively covers our secret war against Yemen (which passes the ongoing relevance test), and chapter 32 covers a particularly horrific episode.
On December 17, 2009 the village al Majalah was bombed, killing dozens of people. Scahill notes how the New York Times reported incorrectly on the bombing, often by relying on anonymous quotes. The Yemeni government claimed credit for the attack, and the Times reported that Yemen's forces had (among other targets) "struck militant bases in Abyan, a lawless area in the south of the country."
The next day the paper ran an article that consisted almost entirely of unsourced and unverified allegations. In less than 400 words it refers to generic officials four times; American officials twice; Pentagon, military and intelligence officials; officials in Washington; administration officials; and Yemeni officials. And it has precisely one named source: Bryan G. Whitman, a Pentagon spokesman. There is literally a 10:1 ratio of anonymous to on the record sourcing.
A week and a half later the story was still roughly the same: "Yemen escalated its campaign against Al Qaeda with major airstrikes on Dec. 17 and last Thursday that killed more than 60 militants. American officials have been coy about the role of the United States in the strikes, saying that they have provided intelligence and 'firepower' for the efforts." In early January the Times still reported it was a Yemeni operation, but now noted many innocents were killed in addition to (or instead of) the 34 militants reported the prior month:
Yemeni officials, in two major strikes against Qaeda targets in December, first said that they had killed Mr. Awlaki, but he later spoke to Mr. Shaea to prove that he was alive, as other key leaders seem to be. But dozens of Qaeda family members and local residents were killed, increasing antigovernment sentiment.1
Then in August the paper quoted an anonymous (natch) Yemeni official reporting no militants and forty four civilians killed in the attack, now attributed to cluster bombs launched by an offshore Navy ship.
So in just under eight months the story from the paper went, with no explanation, from a Yemeni air and ground attack that killed 34 militants2 to an American cluster bomb that killed dozens of civilians (which, in the Times' sedate language, increased antigovernment sentiment). Yet the original articles do not reflect that jarring reversal - as of this writing they remain uncorrected, available to misinform those who come across them.
It would be nice to know how the story traveled so far so quickly, or at least that it had. Maybe some of the many anonymous sources who got the story wrong could be named in an update as well. Unlike a generation ago, articles that are a few years old are not destined to be interred in some journalistic catacomb to only rarely be visited by the odd archaeologist. Books and other projects with broader scope, produced over months or years, will continue to drive traffic to them.
Readers may not take those stories to be an initial snapshot of the event but a living document; not a first draft but a primary source. Accuracy should matter not just on initial publication but after it as well. Perhaps for a handful of the biggest issues, newspapers could spare some resources to make sure their electronic reporting reflects the truth as our understanding of it changes.
1. The wording here is really awkward: "Yemeni officials, in two major strikes against Qaeda targets in December, first said that they had killed Mr. Awlaki, but he later spoke" etc. The "two major strikes" part doesn't assign responsibility for either strike. Mentioning Yemeni officials first would seem to lead the casual reader to the conclusion that it was a Yemeni operation, though.
2. News stories that present a version of events without challenge are functionally endorsing that narrative. The initial report that the attack "kill[ed] at least 34 militants in the broadest attack on the terrorist group here in years, Yemeni officials said." With no counter claim about who was killed, the Times is implicitly crediting that version. Also, there is something more authoritative about printing the allegation first and the attribution second: The attack killed 34 militants, officials said. And your mother is a cheap slut, an insider added.