Gelf Magazine - Looking over the overlooked

Media

May 19, 2005

What We Can Learn From Michelle Delio

Lessons from a journalism scandal on color quotes, note-taking, and credibility problems.

Carl Bialik

Wired News has corrected 24 of Michelle Delio's articles and TechnologyReview.com has pulled all 10 of her articles for the site and published the full results of its inquiry into her work. Delio, a freelance tech journalist, repeatedly cited unverifiable sources that couldn't be found by the inquiries. She wasn't able to supply either probe with information to verify most of the sources.

Sandwiched as it was between a half-dozen other journalism slip-ups, the Delio story came and went quickly and her 15 minutes of infamy appear to be just about over. But there are important lessons and remaining questions worth exploring:

Color Conundrum

Wired News left the affected 24 articles mostly intact. It could do so because many of the quotes were pure color. They could vanish without altering the articles' basic structure. Editors love such quotes and writers dread them: They don't convey any useful information for readers, but instead colorfully reflect a theme of the article, preferably in a folksy way. Gelf has nothing against color in general—anecdotes and detail can add insight and make articles worth reading. At issue here are color quotes in stories that aren't intrinsically colorful and don't involve on-the-scene reporting.

CORRECTION, 6/1: Gelf initially wrote, "Wired News yanked the suspect quotes while keeping much of the affected 24 articles." In fact, Wired News has only pulled quotes from one source who denies having spoken with Delio. The rest of the quotes, from sources who couldn't be located, have been left in the articles, with this caveat: "Wired News has been unable to confirm some sources for a number of stories written by this author. If you have any information about sources cited in this article, please send an e-mail to sourceinfo[at]wired.com." Wired News then lists the missing sources for each article.

In some cases Delio's colorful sources couldn't be found. (Delio's phone was busy Thursday on several calls; she hasn't replied to Gelf's repeated requests for comments over the last two months, after granting an initial, brief interview.) In others, though, she employed techniques that other reporters on deadline might have considered. Consider this passage from the Wired News inquiry by Adam Penenberg, a columnist for the site and a New York University journalism professor who in 1998 exposed New Republic fabulist Stephen Glass:

I asked Delio how she had met Ed McCormick, whom she quoted in the article about Kenyan Masai tribesmen giving the United States a gift of 14 cows, and she said that while she was working on the story, McCormick was doing construction outside her building. Realizing she needed a quote, she told him about the cows and asked what he thought about it. According to Delio, he said, "The cows are the most amazing gift we received—I mean, who else sent cows? If those guys wanted us to have jewelry, they would have sent it. They wanted us to have cows. We should take the cows and raise them on a nice farm upstate and then send the cow puppies back to them some day." But McCormick had known nothing of the Masai gift before Delio told him about it, and she didn't see anything wrong with this.

Or this passage from an email exchange between Delio and Susan Rasky, a Berkeley journalism professor who conducted the Technology Review inquiry.

Rasky: In Rage Against the Machines, how and when did you contact Manhattan graphic designer Jim Heedles about computer rage? Does he live in Manhattan? Does he work for a firm or independently? Do you have his contact information?

Delio: I sent an email to about a dozen friends asking for stories about computer 'abuse' and requesting they forward the email along to their friends if they wanted to. I got maybe a couple of dozen replies, opted to use Heedles for no particular reason, there were several similar tales among the emails. I did not retain these emails, as I thought I wouldn't need them after the story ran. Given that it was a light-hearted supporting quote in the midst of an obviously humorous story I didn't ask Heedles where he worked or lived. I did vet the professor whose comments are central to the story.

Tech Review's Inquiry


Tech Review waited until nearly a month after publishing Rasky's initial report before publishing her appendices (linked from here)—which are comprised of the email exchange with Delio and detailed accounts of Rasky's students inquiries into specific articles, including the assertion from Delio that she inadvertently fabricated one quote. Gelf asked Technology Review editor Jason Pontin why the appendices were published so much later. He replied,

My quibbles were simple: although this there is a fetish in Internet publishing at the moment for full "transparency," the appendices were not originally intended for publication. The "product" was the independent report which Rasky provided us, and which we published in April. It would be akin, I worried, to printing all the interview notes that resulted in a story, and no service to our readers. But while the 1-page Rasky report was very thorough, it left ambiguous what was not ambiguous in the appendices (the question: did Delio invent quotations?) and the appendices also had this virtue: they showed how ridiculous Delio's excuses really were.


It doesn't excuse Delio's poor sourcing to wonder if journalism wouldn't be better off without color quotes entirely. (Though Delio sought to make the distinction in an interview with the Associated Press (via Washington Post), saying, "I don't understand why my credibility and career is now hanging solely on finding minor sources that contributed color quotes to stories I filed months and years ago." In response, Penenberg told Gelf in an interview last week, "I don't make a distinction [between color quotes and others]. Either a story is accurate or it's not.") The quotes take time to gather, take up space that could otherwise be devoted to more-useful material, and are often a way for the writer to inject his or her opinion without directly stating it, by putting it in the mouth of someone else.

Storage Problems

Reading Delio's responses to Penenberg and Rasky, what's striking is how she keeps falling back on incomplete document storage to explain why she can't help them find the questionable sources. Penenberg wrote:

Delio came to my office at NYU for a meeting on Tuesday, March 29, but did not carry her laptop, which I had requested she bring (she said it was broken) nor provide any additional contact information. She volunteered that she has "obviously been negligent" with keeping proper records of sources and perhaps "negligent in getting full backgrounder on every single source" she's ever used. The reason she didn't have any more contact information was that she kept e-mail interviews "for about a month" but then assumed that if no one wrote in to complain, then the piece was OK and there was no need to keep documentation. She also said as a rule she didn't keep notes or source lists.

And in her email to Rasky, Delio wrote:

As I've explained to the absolute horror of others who have been reviewing my stories, I don't retain emails from sources that I don't intend to call on again for comments. And the vast majority of my interviews are done by email. In retrospect, especially given my current situation, I realize that I should have simply dumped every single email pertaining in any way to every story onto some dark corner of my hard drive or—as one reporter I checked with suggested—pasted them into the copy document and burned them onto a CD. But until now I've been a pretty ruthless pruner of documents and data. I typically get a lot of emails in response to my wired news stories, and if the mail comes from someone who is offering to be a source, or who has made an interesting comment about the story I move those messages over into a temporary 'holding' file which is futher subdivided. When I need a quick comment I check this file. Unless a new-to-me source gives me some great insights I tend to use them once (like any reporter I also have my 'usual' sources who I count on) and I don't retain the emails past the point where I assume any questions that might be raised regarding a story would have already been asked. I never, even in my worst nightmares, thought that I'd need to go back and 'resurrect' sources from stories that had run months ago.

There is no set journalistic standard on how long to keep notes. At some publications, the conventional wisdom is that as long as a reporter is consistent in how long she keeps her notes, it doesn't matter how long that is. That's sensible when dealing with the clutter of physical notes. But for reporters like Delio who cover technology, usually from the comfort of offices and homes, that standard no longer suffices. Reporters should be expected to keep computer records of their sources for some significant length of time—a year, perhaps—so that the lack of notes is no longer a viable crutch for bad journalism. Hard-drive space is cheap, one-gigabyte email accounts are free, and it's just as fast to archive as to delete.

Reputational Challenges

It was striking how nonchalant some of Delio's sources seemed to be about problems they found in her articles. One of them emailed her after the article came out: "I'm quoted in it as saying something or other about Kolin, but I had no idea this article was even being written...was this quote taken from me at the gallery? I don't really mind in this case, because I probably would've said the quote anyway, but I would appreciate it next time if you let me know I'm being quoted in an article. Could you tell me where you got the quote?"

Remarkably, Delio responded to him by seeming to put the onus on him for not being available for her to interview: "I tried to contact the gallery half a dozen times last week, sent 4 emails and made 2 phone calls, in an attempt to get in touch with you to let you know that we were doing the article and to see if you had further comments. Unfortunately not one of the emails or calls was returned. I googled and ran various other searches and couldn't locate your contact info on line either."

Later, he found out the article was removed from the Tech Review site, and wrote to her, "I do hope that the article is back up there soon, because I think it's great."

It's reminiscent of how most of the people whose quotes Jayson Blair doctored never bothered to contact the New York Times. Apparently, some people's expectations of reporters are so low that a favorable article with an invented quote is considered, on balance, a good thing.

Delio explained her faulty quote to Rasky this way: "I made a stupid mistake here. I roughed out a quote based on a sheaf of material I read on Joe's art project. It was basically a placeholder that I could write around. I called and emailed his gallery several times to get his contact info and never got a reply. I evidently didn't take out the quote or during the edit process maybe I hit 'accept change' rather than 'reject change' in MS word. I was sure I'd pulled it."

For another article, military sources were less pleased by what they perceived as a misrepresentation of their conversation with Delio, according to Raksy's report. But after contacting Delio, they "did not follow up or demand a retraction."

Undetectable?

Because of the measured reaction of many of Delio's sources, it's possible that she could have continued to practice the same negligent sourcing for some time without editors ever noticing. What ultimately sparked the inquiries into her work was an article critical of a major company built around one unnamed and questionable source—in short, the kind of article most likely to gain attention. It was called "Carly's Way," and it relied solely on G.S., a Hungarian engineer purportedly working at Hewlett-Packard. The company complained that G.S. couldn't possibly exist, the story unraveled and Technology Review and Wired News began their inquiries.

Gelf asked Wired News Managing Editor Marty Cortinas how long Delio might have been able to continue her sourcing practices. He replied by email, "Potentially forever. Or at least until all daily publications backtrack and contact all sources for all stories a second time before publication. That's not going to happen. There are many factors in play, but time constraints and the need to trust your reporters are the two biggest ones here."

Gelf also asked Pontin about this. He replied:

Delio deliberately chose a field without proper fact checking (online journalism), and limited her journalism mostly to stories that no one had much of an economic stake in (culture and technology), and chose to limit her fabrications to inventing people whose existence could not easily be disaproven. It's not true that her fabrications were limited to "color quotes,"—but it probably is true that most of her fabricated quotations play around the edges of stories that were already on the public record. When Delio did take on a really big fiction—like the infamous Chinese cyberwar story—it as always, again, about something that could not easily be checked, and where no one had much of an incentive to challenge her. If you go back through the bulletin boards and blogs since 2000 (see techdirt, for instance), you'll see that many people had questions about her reporting for a long time—but no one chose to complain to Wired because their expectations of technology journalists were so low. [Editor's note: Pontin later clarified that he doesn't directly know whether Wired News had gotten complaints about Delio's reporting.]

This raises the question of whether there are other Michelle Delios still in business. Taking shortcuts on color quotes and not fully vetting sources are practices that aren't necessarily motivated by the thrill of getting away with it—as may have been the case with famed fabulists Jayson Blair and Stephen Glass. Instead, there's a straightforward economic argument for these kind of journalistic shortcuts: You can produce stories faster if you don't dwell every detail, and you'll have a good chance of getting away with it. If you're a staff writer, that means more job security; if you're freelancer, more paychecks. And there are plenty of areas outside online tech reporting that are vulnerable. The fabulist has a decent chance of not getting caught as long as most of these four factors are present:

•A policy of trust-the-reporter, which is to say, no fact-checking in advance. This isn't just the rule at online publications, but at newspapers.
•Sources who aren't used to dealing with the press, or for some other reason are unlikely to complain persistently to editors.
•Article contents that are difficult to refute.
•A moderate audience, which means fewer people who might spot the problems.

Penenberg, when asked by Gelf if he thinks there might be undetected fabulists, replied via email, "Given the sheer number of journalism scandals, it's a fair bet there are a number of reporters out there making things up. There's no way to know how many, though."

Policy Changes

It happened at the Times after Blair and at USA Today after Jack Kelley, and now it's happening in the post-Delio era at Wired News and Technology Review. Both of the tech publications reconsidered and revamped their policies after being stung. The rash of journalistic scandals appears to be provoking reaction mostly on a case-by-case basis; most media outlets appear content to wait until struck with their own embarrassment before tightening the rules.

Wired News said it would require freelancers to submit contact information for all named sources, and restrict use of unnamed sources. Tech Review's Pontin, meanwhile, told Gelf the magazine would be more choosy about using anonymous sources in online articles and would submit unnamed online sources to the same scrutiny that they now receive for print-edition articles.

Why hasn't there been a broader, industrywide re-examination of standards and practices? "Not all newsrooms work the same way," Cortinas told Gelf. "Problems in the production of copy don't necessarily translate in a neat, simple way from publication to publication." In a later email, he pointed out that no policy change would suffice to catch a determined faker, short of fact-checking every piece—an expense almost no publications other than weekly glossies choose to make (and one that didn't spare the New Republic the Stephen Glass disaster). "It comes down to trust, and unfortunately there are always going to be people willing to exploit that trust," Cortinas said.

That's true, but tougher sourcing rules can make it tougher to exploit the trust, and when something gets harder to do, fewer people do it.

Gelf isn't suggesting a one-size-fits-all sweep of policy changes; Cortinas is right that each publication is different. At Gelf, we trust our reporters. But each publication should rethink whether its current policies and practices would make things tough for a fabulist in their midst (for example, occasionally Googling unfamiliar sources, and bringing more skepticism to editing). Just in the last week there have been three more embarrassments:

•Gannett's Courier-Post, of Camden, N.J., found that a staff writer was using passages from other Gannett newspapers without attribution. That writer "is no longer employed by the Courier-Post," the newspaper said.
•A Sacramento Bee columnist resigned amid an internal inquiry that couldn't verify several of her sources. In a follow-up interview with the Sacramento News & Review, the columnist "wouldn't talk specifics about her columns" but said, "There's a witch hunt going on in American journalism."
•A Detroit Free Press inquiry found one of its columnists "at times has used quotes from newspapers, TV programs or other publications without indicating that he did not gather the material himself, in violation of Free Press rules on crediting sources." This was somehow deemed good news; the headline said the probe "shows no pattern of deception."

It's also possible that some publications are catching problem cases through unannounced, internal policy changes and inquiries. But they'd do better to announce their new measures so others can learn from them. (Also, there's no indication that more journalists are sloppy or fraudulent today than in the past. Thanks to the internet, articles are more widely distributed, readers can email corrections, and editors can more easily look for plagiarism and questionable sources.)

It's unlikely that publications will go so far as to institute advance fact-checking, which is expensive and slow. As Pontin told Gelf, that's almost certainly not going to happen online: "Most sites are still wildly underfunded in terms of their editorial budgets. This is because while online advertising is growing much, much more quickly than print advertising (which is, indeed, in many markets declining), the actual sums are still quite small. Technologyreview.com, for instance, accounts for less than 15 percent of our total revenues. Since most sites are run as separate businesses in publishing companies, they tend to be allocated paltry monies—certainly not enough for a dedicated fact-checker."

This discrepancy helps explain how the story that started it all, "Carly's Way," got published. As Pontin told Gelf:

Technology Review must take full responsibility for not fact checking the story before publication. The story was originally intended for a "By Invite" column in the print magazine [here's the latest one, as an example]. I killed it because I thought the tone of the piece was unpleasant, and because I didn't want to publish an anonymous story. Had the story continued to move through the print process, it would have been fact checked by our dedicated fact checker—and we would have caught the fabrication. The story would never have been published. But when we decided to run the story online, I instructed Brad King, our Web editor, to carefully fact check the story before publication. He failed to do so, because (he says) a print and online editor mean different things by "fact checking." Online, we check basic matters of fact—but otherwise trust to the honesty and accuracy of the reporters.

Delio cited this lack of fact-checking in her defense, in her email to Rasky:

I assume you know that G.S. provided a bio, at Jason Pontin's specific request, which I forwarded to Pontin when I submitted the story on 2/14. I was told that the bio would be vetted—Pontin made a point of telling me that. I can forward the email to you if you need it. The bio listed G.S's education and current professional affiliations. I found out the bio hadn't been verified only after the story ran two weeks later and HP questioned it. I don't always do backgrounder on sources, but in a single-source, anonymous source piece attacking a prominent company and businessperson, I certainly would have done a quick check of the bio info had I thought Tech Review didn't plan to do so. I had told G.S. the bio would be vetted, so he may have assumed we would have figured out it was a scam before publishing. Again, that's only (and obviously) a guess—I have no idea what he was thinking.

Of course, had "Carly's Way" been killed by King, it never would have caught the attention of H-P, and Delio might still be a contributor at Tech Review.

Dreary, Expensive Work

Here's another reason for publications to take prophylactic action rather than waiting for their own embarrassment: Rereporting questionable stories is dreary and expensive work. Tech Review paid Rasky $5,000. And she told Gelf that it wasn't an experience she enjoyed much. "I don't think it's fulfilling at all. ... I would not particularly want to undertake this again." She added, "I was very conscious as we were doing this that we were dealing with somebody's reputation, and that what we said and how we said it would matter a lot to this woman's career and her livelihood."
Gelf got a taste of what's involved in our efforts (here and here) to confirm the sourcing in several of Delio's articles. There was no pleasure in the process.

Penenberg's work in exposing a faked Stephen Glass article was dramatized as glamorous detective work in the film Shattered Glass, thanks to the soundtrack and the presence of Rosario Dawson. Yet in an earlier interview, Penenberg told Gelf he took no pleasure in doing such work. Sadly, though, it had become necessary: "We have an epidemic of plagiarism, fabrication and—I don't know what you call it—reporters or columnists claiming to be places where they haven't been."

Career Path

Blair and Glass, who admitted to widespread deception, probably will be out of mainstream journalism forever. But what happens to reporters in the grey area now inhabited by Delio?

In the email interview with Rasky, Delio indicated that she knew her career might be over. She wrote at one point, "Obviously, if I'm allowed to continue on as a journalist, I'll keep a file on everyone who is quoted in any of my stories."

It's pretty clear Delio will never work for Tech Review again, at least not while Pontin is in charge. In emails to Gelf, he referred to being "extraordinarily angry" about having been "defrauded."

But Wired News isn't closing the door. Cortinas told Gelf, "We haven't made any official proclamation on using Delio in the future. We don't plan on assigning her anything, but we may revisit that at some point. I think you should always be wary of absolutes, and 'never' is a big word."

David Goldenberg contributed to this article.

Carl Bialik

Carl Bialik, a co-founder of Gelf, is a writer for FiveThirtyEight.







Post a comment

Comment Rules

The following HTML is allowed in comments:
Bold: <b>Text</b>
Italic: <i>Text</i>
Link:
<a href="URL">Text</a>

Comments

- Media
- posted on May 21, 05
Sam

What is Gelf's policy on disclosing how an interview was conducted? I'm not sure that this is an ethical issue, but when I read a quotation in a piece of journalism, I assume that the interview was conducted in person. I notice that Gelf often mentions that sources replied by email. Are all email responses labelled this way? Can I assume that unlabeled quotations were spoken directly? What is the policy for telephone interviews? This may seem like a small issue, but to me, a spoken answer has a different character than a written one.

- Media
- posted on May 22, 05
Carl

Sam, good question. Most of our interviews are conducted by email or telephone. We'll specify when it's in email or in person; if we don't specify than it was by telephone. That seems to be pretty standard in newspapers and in magazine stories that aren't based on on-the-scene reporting. If you see a quote from a defense analyst or expert professor in the newspaper, it was generally based on a telephone interview.
I'd like to hear your thoughts on whether this is appropriate disclosure, and more on how you read email quotes differently from others. To me, in-person quotes offer advantages of immediacy and color, but for an article like this one, email is preferable because it gives the respondent time to carefully and thoroughly answer complex questions.

- Media
- posted on May 24, 05
hostile17

"if I'm allowed to continue on as a journalist"? "Continue ON"? She should be severely reprimanded just for that sentence...

- Media
- posted on Jun 01, 05
robert

"Wired News yanked the suspect quotes while keeping much of the affected 24 articles. "

I looked at one of Wired News' delio articles ( http://www.wired.com/news/technology/0,1282,65906,00.html ) and it seems as if they didn't remove the suspected quotes. They mentioned the suspected sources, but the quotes remained in the text if the questioned sources could neither be confirmed nor denied. You do check your own copy for accuracy, right Gelf?

- Media
- posted on Jun 01, 05
Carl

Robert, you're right. Wired News also brought that to our attention. Thanks for letting us know. We've posted a correction.

- Media
- posted on Jun 20, 05
David Arv Bragi

First, I should disclose that I am an old friend and former colleague of Michelle Delio. My comments, however, are of a more general nature. One of the main reasons that freelance reporters are getting more sloppy is that they literally do not have the time to do proper research. Pay rates have declined to the point that a feature article that should normally take 20 to 40 hours to complete might pay only a couple of hundred dollars. This is why I left the freelancing profession. I could not earn a living writing the news unless I cut editorial corners. If every reporter in the country did as I did, there would be very few left in the industry.

- Media
- posted on Nov 28, 07
Anders Lautrup-Larsen

I know Michelle Delio from a series of interviews I did to her back in 2002.
My expirience was that she did a very deep and prof. research hearing all parties, and getting the e-mail and phone numbers etc. as refferences(very different attitude then other leading medias arround the globe)

- Media
- posted on Jan 02, 16
Meeka

What do you expect from a family of money hungry cheaters. From her spouse's failed law suit, brother in laws work ethic's and even what hergood old grandma did to her family.


Article by Carl Bialik

Carl Bialik, a co-founder of Gelf, is a writer for FiveThirtyEight.

Learn more about this author






Newsletter

Hate to miss out? Enter your email for occasional Gelf news flashes.

Merch

Gelf t-shirt

The picture is on the front of the shirt, the words are on the back. You can be in between.