If a tornado were to roll past the Statue of Liberty — and past 8 million of the most connected people in the world — there would be a corresponding storm of photos and eyewitness accounts in social media, right?
Right. So when a single photo surfaced on Twitter last week amid all the images of downed trees in New York, its uniqueness should have been a warning sign.
Reporters are turning to social media for first-person accounts and images of news events such as the recent hostage situation at the Discovery Channel headquarters near Washington, D.C. And we all know about the photo of the plane in the Hudson River. So it made sense for journalists to turn to the crowd to document a freak storm in the middle of the city.
It turned out that one of the photos they picked — an image of a twister going past the Statue of Liberty — was 34 years old. The photo was posted by Gawker, The Village Voice and Time, and it spread across Twitter like — well, fast.
“When it comes to something like the tornado,” Adrian Chen, who posted the photo on Gawker, told me via e-mail, “you’ve got to have something up quickly, and you’re pressed hard by the knowledge that thousands of people are tweeting about it, basically ‘scooping’ you in real time. So you have to use pictures from Twitter. “
But in a post reporting that the photo wasn’t authentic, Huffington Post Traffic and Trends Editor Craig Kanalley wrote, “Journalists must do a better job vetting the information and multimedia that flows from social networks. We are making it far too easy for folks to dupe us. Especially in situations of chaos, when we’re most vulnerable.”
I find no joy in highlighting the mistakes of journalists working on deadline. But journalists and news consumers are going to see more of these images and accounts passed through social media, so let’s step back and see what happened here.
When presented with the incredible, apply some skepticism.
Kanalley (whom I know professionally and has contributed to Poynter Online) told me that he was intrigued when he saw the photo posted on Time’s website. “My very first instinct was, I have to get this photo on Huffington Post — if it was legitimate,” he said.
But he was suspicious, too. Even though people were abuzz with how violent the storm was — and the National Weather Service later confirmed that two tornadoes had struck the city — this seemed too good to be true.
Within a few searches on Google, he found the same photo on a National Weather Service site — from 1976, not 2010. Thinking he may be misinterpreting what he saw there, he did some more searching and found the tornado referenced elsewhere, including in a book about the 1976 tornado season and in a recent Farmer’s Almanac tweet.
Kanalley posted a story describing Time’s mistake (at the time, he only knew that Time had posted the photo) and tweeted Time to inform the staff that the photo wasn’t from the storm. Time pulled the image and posted a correction. Gawker and The Village Voice issued mea culpa’s too, though not as prominently.
Let’s say this photo hadn’t been so easy to find online. What are the chances that just one person would have captured an image like this? In major cities, people are taking photos and video of all kinds of things, practically all the time. Consider all the images we saw of the planes striking the World Trade Center.
If we don’t apply a critical eye, our readers will.
Earlier this summer, when the Chicago Tribune posted a staff photograph of lightning striking two skyscrapers at the same time, many readers responded with skepticism. How did they know the photographer hadn’t used a long exposure to make the strikes appear simultaneous, the same way they capture those beautiful fireworks photos?
Photo editors responded that another Trib photographer had captured almost the same shot, and that a non-staffer had captured incredible (my description, not theirs) video of the strikes. Three images of something that lasted just a split second.
“Next time, I’ll do a bit more digging around Twitter to verify the authenticity of a photo I find there, maybe by contacting the person who posted it, or seeing if others had posted similar pictures,” Chen told me. “And I imagine I’ll opt for a more plausible — if less dramatic — photo.”
Check your source.
One of the first things Kanalley did was visit the Twitter page for “TheDaveCarlson,” the person identified as the source of the photo. Kanalley told me many of the tweets were jokes, not news items, which added to his initial suspicion. (None of the tweets are visible to the general public now because the user changed his account to “protected.”)
Moreover, Kanalley noticed that although the photo was grainy — perhaps taken with a cell phone — the tweet had been posted via Twitter’s website, not a mobile phone application. In a breaking news event like this, he would expect someone to post the photo directly from a phone.
In 140 characters, it’s hard to read between the lines.
Time’s initial correction, according to Kanalley’s post, said, “Twitter, and Dave Carlson, have led us astray. The image is from 1976… “
Yet Carlson told Kanalley that he posted it “merely to mess with friends and when asked I told them it was fake. A friend told me TIME posted and I couldn’t believe it. I wasn’t asked by him [the reporter] if the photo was real or not.”
It’s hard to pick up on subtext in e-mails; it’s even harder to do so on Twitter, where earnest, nutjob and ironic tweets all look the same — especially to strangers.
In response to Time’s statement that it had been led astray, Alex Howard (@digiphile) urged the magazine not to blame the messenger. “Is Gmail at fault if users [forward] flawed stories?” he tweeted. “People duped, not medium.”
Pack journalism isn’t the same as verification.
The Village Voice posted the image under the sentence, “If this is real, [the storm] is more insane than originally thought.” It followed with this: “Further support that the above photo is real. Time posted and tweeted it.”
Likewise, Chen told me that he came across the image on The Village Voice’s site and noted that it had been picked up by Time.
“Seeing it on Time, longtime champions of photojournalism, I wrongly assumed it was legit,” he said. “Since it was an amazing picture, I decided to use it for the post” on Gawker.
When journalists are working in a breaking situation, they understandably look at the competition. But they should remember that the competition is looking back at them. Having another news outlet follow your lead doesn’t mean you’re right; it could mean you’re both making the same mistake.
“The biggest lesson I learned is that seeing an established name pick something up from Twitter isn’t an automatic seal of authenticity,” Chen said.
If you try to make something go viral, try just as hard to follow up with the cure.
Kanalley noted in his story that Time tweeted its post “to StumbleUpon no less, to give it more viral potential.” Later, Time corrected the post and deleted the tweet — but the magazine never sent a follow-up tweet correcting the information, Kanalley said. “It continued to go viral even after that,” Kanalley said.
News orgs should spend just as much time trying to draw attention to corrections on social media as they do publicizing their original, erroneous information. It’s hard to chase down every retweet — especially when you have millions of followers like Time — but isn’t that part of our responsibility to our audience? Considering that it’s easy to track retweets, it’s possible to send @responses to everyone who retweeted the erroneous information.
In response to my question about what happened, Time spokesman Daniel Kile said via e-mail, “TIME is a 24/7 news site that is updated continuously in real time by a large staff of editors and reporters. As soon as the error came to our attention, it was corrected immediately and prominently.”
As I was finishing this post Monday afternoon, I saw an e-mail alert from CNN reporting that economists now say the recession ended in June 2009. But I didn’t read carefully and griped on Twitter, “Why did it take a yr for economists to figure out that we were in a recession and when it began, but just 3 mos to conclude that it’s over?”
I soon received a gracious e-mail from Jim Naughton, past president of Poynter, who told me that it was June of 2009, not 2010. Lucky for me, I have far, far fewer than the 2.2 million followers that Time has, so people just ignored me and the sheepish correction I tweeted.
Lesson learned: Don’t spout off on Twitter.