Surviving the Grind of Fact-Checking

A balled-up piece of paper with the word TRUE typed on it repeatedly sits on a piece of paper with the word FALSE written on it over and over again.


It was a low moment, the kind most writers remember for a long time. I was sitting at my desk, wrestling with a tight deadline, when I received an email from one of my editors. A reader had just written in challenging the accuracy of a description I had given in a recent story. Could I look into it, my editor asked? I began scouring my research and realized to my chagrin that the reader was right: I had slipped up. The following day I sat down and wrote a correction, which ran in the next issue of the magazine.

Even the most careful writers get caught out committing an error from time to time. Editors seldom rank these blunders near the cardinal sins of plagiarism or fictionalization, but frequent errors or a cavalier attitude towards accuracy can quickly erode the credibility of a writer and the reputation of a magazine. To guard against this, large publications often hire fact-checkers to root out mistakes before articles see the light of day.

Many science writers I know have a kind of love-hate relationship with fact-checkers. In principle, most appreciate that someone is looking after their backs, but they can’t help seeing the process itself as a little humiliating and painful, right up there with having a root canal or a colonoscopy. “I really hate it with all my heart and soul,” says veteran science writer Ann Finkbeiner. “It’s everything I left grade school to get away from.”

Many large-circulation magazines insist on fact-checking, however, writing it into their letters of assignment. National Geographic, for example, requires writers to provide a manuscript copy that includes “the annotated list of publications, addresses and phone numbers of people and institutions mentioned in the text and other sources of information,” and to submit all the source material they collected during the assignment. The magazine then assigns a fact-checker to go through the story with a fine-tooth comb, checking each fact against the backup material to ensure that the writer has gotten even tiny details right.

I’ve been through this kind of grueling fact-checking many times, yet I still feel a little anxious every time a fact-checker calls. But I have learned over the years that a writer can keep a fact-checker’s queries to a minimum by making the best and most detailed annotated copy possible.


Keeping Track of Source Material

Zach Zorich, a former fact-checker for Discover magazine and now a freelance science journalist, says the best time to start thinking about the annotated copy is when you begin researching a story. Zorich prints out the journal articles he comes across that contain valuable source material and then stores these print versions in manila folders in his office. He admits that this method is a little old-fashioned, but says, “I prefer having a hard copy of the thing. Maybe it’s the instinct I had when I was a fact-checker, that if I can’t put my hands on a piece of information, it doesn’t exist.”

I started out using that exact method, but have now switched to digitally archiving my source material. I usually download relevant journal articles as PDFs or use a Microsoft program called OneNote for taking screen shots of interesting web pages and storing them on my computer. Often I will do this as soon as I find something interesting, rather than waiting until after editing is done and trying to backtrack. Companies sometimes post information about their activities that is later considered damaging, and when that happens, a webmaster can quickly delete the incriminating page. But with the original printed and stored in a manila folder or preserved in a program like OneNote, a writer has solid evidence to pass on to a fact-checker.

The best written sources for a science story, by the way, are always primary sources, such as a researcher’s own published papers. Using information derived from secondary sources such as a Wikipedia entry or a newspaper story reporting on a scientific discovery won’t cut the mustard when it comes to fact-checking. You could just be recycling someone else’s errors.


Marking Up a Story

Once a story has been edited and accepted, the editor generally lets the writer know when to submit the annotated copy and accompanying backup material. (There’s little point in going to the trouble of annotation before then, because an editor may have to cut a long passage at the last minute in order to make a story fit the available space.) As soon as I am given the green light, I try to get down to annotating as quickly as I can, since fact-checkers often work on tight deadlines.

An annotated copy is just what it sounds like. It contains footnotes or embedded comments that tell a fact-checker where the writer found each piece of information, whether from human or written sources. Word processors such as Microsoft Word make the technical process of inserting footnotes easy; the hard part is remembering where each discrete piece of information came from.

I asked David Lande, a senior researcher and fact-checker at National Geographic, what really good annotated copy looks like, and he sent me what he called “a shining example,” the backup that Joel Bourne submitted for his April 2010 story “California’s Pipe Dream.” The article was some 2,700 words long, and it contained 86 footnotes—roughly one per sentence. These annotations provided Lande with all the info he required to check the facts: the name, phone number, and email address for each human source; links to all written online sources; and the titles of PDFs and books consulted by the writer. The author then emailed the PDFs and even sent entire books to back up his annotations. For the sake of expediency, Lande told me in an email, he appreciates it if the author sends entire books, which he always returns.

There seem to be no hard-and-fast rules when it comes to the submission of reporting notes, interview tapes, and transcripts, however, and a writer is best advised to ask the fact-checker about preferences. Lande, for example, is happy to receive digital interview recordings or transcripts of interviews. Though he’s also glad to receive handwritten notes if that’s all a writer has, he says he is less keen about poring over a reporter’s handwriting. “The adventurers come back with little pocket notebooks [that] they provide, but it’s often hard to decipher the notes,” he says.


Thin Annotation Is More Than Rude—It’s Risky

Over the years, Lande has seen just about everything from writers. He told me in our email exchanges that he was in the midst of a fact-checking nightmare, working on a feature of about the same length as Joel Bourne’s story, but which arrived with “a grand total of 10 footnotes.” The writer, Lande noted, had previously written at length on the subject, but the flimsy annotation meant the fact-checker had to do far more work than he had bargained for in chasing down sources.

Moreover, writers who go too light on annotation risk a more serious problem. Often, Zorich says, fact-checkers who don’t get the information they need from a writer have little choice but to start tracking down potential sources on their own and checking out the story with a new cast of researchers. Rarely are writers happy with the results, because the new sources have different ideas and points of view than those of the original contacts. “That’s where, as a writer, you’re going to get your article severely tweaked,” Zorich notes.

Writers can save themselves a lot of grief by working with, rather than against, a fact-checker. After all, the fact-checker and writer are ultimately on the same team. “In science writing,” Finkbeiner says, “you are just a fraud if you don’t get the details right.”


Heather Pringle Courtesy of Heather Pringle

Heather Pringle is a contributing correspondent at Science and a contributing editor at Archaeology. She has written four books, including The Mummy Congress: Science, Obsession, and the Everlasting Dead. Follow her at Twitter at @hpringle.

Skip to content