It was a question Keren Landman had gotten before—but this time, it caught her off guard.
Landman, a physician and journalist based in Atlanta, Georgia, was interviewing a fellow medical doctor and occasional freelance writer who made a familiar request: Can you run part of your draft by me?
Not the whole draft, her source was quick to clarify. Just “the actual quotes” that Landman was planning to use, “in case there’s anything embarrassing.”
Landman pushed back, explaining that she wasn’t out to make anyone look bad—and that sharing unpublished material was against her editor’s policy. But her source bristled. “When I’ve been on that side,” Landman recalls her saying, “I feel like you can at least share the language.”
Landman and her source eventually reached an understanding that didn’t involve the exchange of unpublished copy—and both were happy with the final piece. But the sticky situation they’d run up against is common in science journalism: sources attempting to invite themselves into the crafting of a story in progress. (An informal survey of the 12 journalists interviewed for this piece suggests that these situations arise with about 20 to 40 percent of sources, typically in interviews with academics who don’t have much journalism experience.) This kind of incursion on the reporting process can take many forms, from requests for quote approval, as happened to Landman, to demands to see an entire piece.
At their most extreme, these conversations can even get combative: A source once insisted that he would refuse to go on the record unless I agreed to review my entire piece with him before publication, effectively holding his material hostage. Since that was something I couldn’t agree to, I didn’t include him in my story. Most other instances are considerably more civil—but even when draft requests come with the best of intentions, journalists overwhelmingly agree that they can’t be accommodated without a breach of integrity. (Certain outlets or editors may allow for rare exceptions, but it’s best to default to a stricter protocol.) Given that, navigating these conversations kindly and effectively has become a critical part of the science journalist’s job.
The practice of not sharing drafts “is designed to keep sources from having influence over the journalist and the resulting piece that they write,” says Betsy Mason, a freelance science journalist and editor who lives in California’s Bay Area. But when the question comes up, Mason adds, “don’t be offended. Try to find out why, and try to talk it through with them.… Every time this is asked of you, it’s another opportunity to help a scientist understand journalism better.”
A Very Common Question
Issues surrounding quote approval and draft sharing have arisen at many publications in the past, and are by no means limited to science journalism. After a highly publicized 2012 incident that involved quotes being volleyed back and forth between reporters and White House officials, The New York Times explicitly revised its policies to forbid “after-the-fact quote approval.”
But most scientists don’t know the ropes of journalism, and, despite increasing expectations that researchers participate in public outreach, press training is still absent from most academic curricula. That leaves academics ill equipped to navigate the average interview, where scientific jargon is discouraged, but key journalistic terms like “off the record” or “on background” hold enormous power. Unfortunately, though these terms might sound familiar to many, they aren’t often understood or wielded correctly by scientists, says Steve Brusatte, a paleontologist at the University of Edinburgh who has written extensively for general audiences and is among the most-quoted experts in his field.
A reporter approaching any source for information is also making an inherently intimate request—often for something sensitive, personal, or even incriminating—and it’s understandable that many on the receiving end of such requests are wary. “These are human beings that don’t know you, and don’t have a reason to immediately trust you,” says Akintunde Ahmad of the Columbia Journalism Review.
What’s more, scientists being interviewed are often asked to communicate an enormous body of work—sometimes years in the making—in just a half hour. After spending their entire careers scrutinizing and protecting detailed data that no one else has yet acquired, researchers are understandably wedded to their work, says Jonathan Wosen, a health reporter for the San Diego Union-Tribune. “People take their work very personally.”
Furthermore, Brusatte adds, fast-turnaround news stories may seem especially foreign to researchers who are used to the plodding, months-long peer review process of scientific publication, during which drafts are, as a rule, shuttled back and forth between writers and editors. And for some, fears of factual errors, sensationalized writing, or misrepresentations of opinion have unfortunately been borne out before.
Their Purposes, or Yours?
When scientists and journalists enter into early discussions, a bit of wariness may be understandable—and perhaps even expected. But the path to establishing trust doesn’t require draft sharing, which, even with the best intentions, can open up an ethical can of worms.
As is the case with all journalists, science reporters have a duty to find out “what’s really going on,” Mason says. That means talking to multiple sources, handling inconsistencies, and, perhaps most importantly, remaining skeptical and objective. Allowing a source to infiltrate a writer’s document can compromise that independence.
Erika Check Hayden, director of the Science Communication Program at the University of California, Santa Cruz, draws an analogy to political reporting. A statehouse reporter would never be expected to share a draft with a local lobby group or a government official featured in their story; similar ground rules hold true for scientists as well.
Of course, writers should still contact their sources when necessary to check facts, clarify a confusing statement, or simply follow up with more questions, all of which can be accomplished without sharing exact copy. Journalists worried about the accuracy of something they’ve written can rephrase or summarize bits of text in an email or, if possible, over the phone, says Purbita Saha, a journalist and editor at Popular Science.
That can feel especially intimidating when writing about a subject far outside one’s comfort zone. My background in biology was of little use last year when I wrote a piece that required me to explain thermodynamics. Worried that my interpretation of scientific jargon would introduce errors or confusion, I checked paraphrased pieces of my story with four different experts—two of whom weren’t interviewed for the article—tinkering with my adjectives and metaphors until all my sources had converged on something similar.
Beyond straightforward fact-checks, follow-up conversations with sources can also help with sensitive issues, such as ensuring a source’s anonymity, or avoiding misrepresenting members of a marginalized community, Ahmad says.
Mason recommends keeping in mind a crucial question: Is a follow-up “for my purposes, or theirs?” Clearing a word, phrase, or number with a source can improve a writer’s story by giving it more depth and accuracy. Beyond that, though, a clarity check may turn into an unwanted collaboration, taking editorial control out of the writer’s hands.
How to Say No
When sources request your partial or full draft, there isn’t always time to give them the full breakdown of the importance of journalistic independence. But if the conversation allows, Mason recommends trying to find out where the question is coming from and addressing the interviewee’s concerns as thoroughly as possible. “It’s important to discuss,” she says, to ensure that both parties have a clear-eyed view of this journalistic convention.
Mason also emphasizes that she shares her sources’ commitment to accuracy: “I usually explain that my primary goal, my very top mission, is to get everything precisely right,” she says. “And that if I’m not sure, I’m going to check back with [them].”
Many researchers aren’t aware that some outlets have dedicated fact-checkers who will rigorously parse and vet every sentence in an article to ensure that no errors have been introduced during the reporting process. (Sometimes that even involves recontacting sources and verifying pieces of an article with heavy paraphrasing, just as the original reporter would have done.) Simply explaining that this system exists may bring sources some comfort, says Rachel Gross, a science journalist and editor based in Boston, Massachusetts.
Even if a dedicated fact-checker doesn’t enter the editing pipeline, part of the journalist’s job is also to fact-check their own work; in the absence of that expectation, sources would be right not to trust the reporting process. If a source has consented to being recorded, Mason adds, that’s also a plus, because she can point out that she has a record of all quotable phrases and facts in the exact context they were given.
Just as a scientist wouldn’t want their name attached to an erroneous assertion, these articles also have “your name, and your publication’s name on them,” Gross says. For all involved, “the last thing you want to do is get the facts wrong.”
Several writers also recommend citing a publication’s (or editor’s) policy forbidding draft sharing—something that’s both official-sounding and takes the reins out of the writer’s hands. The San Diego Union-Tribune’s Wosen tries to also put forward a substitute that might allay some of a source’s concerns, saying something like, “I’d be happy to spend a couple minutes after our interview going through some of the main points that you raised.” Even if sources don’t take him up on the offer, Wosen says, they seem to appreciate his commitment to journalistic rigor and accuracy.
Above all, exercise empathy, Ahmad says. “Have compassion. Let people know, ‘I understand how this could be nerve-racking for you.’” It can also be helpful, he adds, to assure people that you have no wish to misrepresent them or the truth.
These conversations can also be preempted by starting an interview with a quick rundown of the journalistic process, and what can and can’t be shared during drafting, Check Hayden says. If any confusion or hesitation persists, journalists can also offer their sources helpful online resources that explain certain conventions, such as this write-up from the American Press Institute.
During Landman’s conversation with her wary source, transparency seemed to work in her favor. She remembers telling her source, “We’re really discouraged from running things by people before publication. I can again reassure you that it’s not my practice to make people look bad.”
None of this is easy, especially when a source really puts the pressure on. “It can definitely be intimidating when a source makes this demand of you,” Saha says. But to serve the reader—to serve the public—it’s worthwhile to take a stand.
“Having this ethical line … is what keeps our practice honest, and helps deliver the best information to readers,” Saha says. “That’s why our industry exists.”
Paraphrasing for Fact-Checking
Below are three examples of passages from three recent articles that Katherine Wu reported. The left-hand column shows the text in Wu’s article before she fact-checked it with her source. The middle column approximates how Wu paraphrased that material when presenting it to her source. The right-hand column shows the final text in the published story.
| Story text before fact-checking | Paraphrasing used in fact-checking | Final text reflecting any fact-checking changes |
| Perhaps more common is partial protection, [source] says. In these cases, a bug might infect a person a second time but struggle to replicate in the body, causing only mild symptoms (or none at all) before it’s purged once more. | In other more common scenarios, a person may have varying degrees of partial protection, making it possible for them to be reinfected, in some cases experiencing symptoms. | People can experience varying shades of partial protection in the wake of an infection, [source] says. In some cases, a bug might infect a person a second time but struggle to replicate in the body, causing only mild symptoms (or none at all) before it’s purged once more. |
| When most mammals detect a viral invader, they’ll battle it with inflammation, a wildfire-like weapon that helps purge the pathogen from the body, but can also wreak havoc on the host’s own tissues. Bats, on the other hand, are able to tamp down these risky responses. Part of this dial-down may involve a repurposing of the molecules they’ve evolved to mitigate the detriments of flight … | Part of the reason bats don’t seem to be as affected by viral infections is because they’re able to tamp down some of the inflammatory responses that might drive disease in other species. This seems to be related to the ways they’ve evolved to cope with the metabolic demands of flight, which might use some of the same molecular machinery. | When most mammals detect a viral invader, they’ll battle it with inflammation, a wildfire-like weapon that helps purge the pathogen from the body, but can also wreak havoc on the host’s own tissues. Bats, on the other hand, are able to tamp down these risky responses. Part of this dial-down may overlap with the ways they mitigate the detriments of flight … |
| Several research teams are designing human vaccines that trigger the production of antibodies that attack SARS-CoV-2’s spike protein—the molecular key the virus uses to unlock and enter human cells. Because the spike protein is crucial for viral infection, it makes an excellent target for a vaccine, says [source]. But [source] also points out that the spike protein, like other parts of the virus, is capable of mutating—something that could compromise the ability of a vaccinated individual to ward off the virus. | Several research teams are now trying to target the spike protein. This is a good option, with the caution that the spike protein can mutate, so we need to be cautious … one potential implication is that we may have to reformulate vaccines every couple years. | No changes |
Katherine J. Wu is a science journalist and Story Collider senior producer who has written for Smithsonian, National Geographic, Popular Science, Undark, and more. She is a TON early-career fellow sponsored by the Burroughs Wellcome Fund and is currently a reporting fellow at The New York Times’ science desk. She holds a PhD in microbiology from Harvard University. Follow her on Twitter @KatherineJWu.
