We Need Investigative Science Journalism But Learning to Investigate Science is Hard-Part 1

[^^”How do you know?”: The question that science journalists must not forget to ask.]

One night about a month ago, I was at a friend’s birthday party, knocking back tequila and rum with assorted MIT-affiliated twentysomethings. Somehow I ended up talking about tardigrades with a post-doc from an  uber-spiffy genetics institute.

15994168283_0b887a34c3_o

[This is what a tardigrade looks like. Photo via Peter Von Bagh]

Tardigrades are a clan of microscopic but thoroughly adorable invertebrates, that recently found themselves at the center of a huge genomics controversy.  The tardigrade genome “kerfluffle” also happened to be one of the stories I wrote about for MIT Science Writing class.

So when the post-doc told tipsy me something to the effect of: “The guy whose tardigrade genome paper got criticized actually came to our institute and gave a talk. We found out later that some of his slides had been plagiarized from a third tardigrade genome group in Japan!” I was pretty appalled.

(Quick Version of the Tardigrade Genome Backstory:  In November, a paper in the highly prestigious Proceedings of the National Academy of Sciences journal announced that they had sequenced a tardigrade genome and found that 1/6th of tardigrades’ genes had come from bacteria.  Bacteria swap genes back and forth amongst themselves all the time, and animals do carry lots of genes from invading viruses that somehow found their way into germ lines.  It’s not impossible for bacterial genes to find their way into animal genomes, but it’s rare.  The 1 in 6 result suggested that tardigrades have a second superpower, which allows them to grab genes from bacteria. If true, that would be a huge f***in’ deal for invertebrate zoologists and geneticists.

But then, a week later, a second tardigrade genome project posted a preprint of their study on BiorXiv, announcing that by their count, tardigrade genome had only an average number of bacterial genes. Many in the genomics community interpreted that to mean the other team’s finding was probably due to contamination.)

Genomics scientists argue about whether some genes are “artifacts” from contamination, rather than actual parts of the genome all the timeThat’s a big part of what genomics is.

However, plagiarism is a cardinal sin in science. So is deliberately overhyping your results, even though that happens all the time.

And the alleged plagiarist was someone I had spoken to.  In fact,  he was someone whose work inspired me to spend multiple hours trying to figure out how to explain it properly in normal people words.  Writing about science is my job, but at the same time, I try to be selective about which scientists receive my time and energy.

In the case of the tardigrade story,  my teacher actually rejected my initial pitch in favor of a story about neurons. But when I called up the neuroscientist, he launched into a 40 minute spiel, without giving me time to ask my own questions, as if I were a stenographer, not a fledgling journalist.  I try to avoid writing about scientists who treat me that way, and besides that, tardigrades are one of my favorite animals. I wanted to write about them.

I had put a lot of time and energy into learning about the tardigrade researcher’s project, the arguments against it, and the context around it.  I hadn’t cheated. I hadn’t slacked. But feeling like I had expended all that energy into writing about work by someone who might have been cheating at science made me feel pretty violated.

“Really?!” I squeaked at my drinking companion.

“Yeah,” he replied. “After the faculty called him on it, he just disappeared…Definitely a sketchy guy.”

I wanted to melt into the floor and disappear myself. By most accounts, I’m a rather promising young science journalist.  I’m supposed to be one of the first lines of defense against falsely scientific hoodwinking and gobbledy-gook. How had I missed the signs that something about that paper wasn’t quite right?

Reporters *Do* Know That Science Has Politics, Economics, & Scandals…

On Tuesday, an article called “How Journalists Can Help Hold Scientists Accountable” from (the excellent) Pacific Standard
landed in the science journalism Twitter steam with a loud SPLASH! 

The article’s author argues that “science journalists tend to position themselves as translators, “that we widely regard information scientists tell us as “unimpeachable fact,” and that science writing still isn’t sure whether its goal is to be journalism or to be PR He also strongly implies that science journalists don’t know that science is “messy”.

To a journalist, those are fighting words.

And a lot of science journalists reacted to it negatively, myself included. Many of those arguments lapsed into #NotAllScienceJournalists-style arguments, but for me, that wasn’t the problem.

In science journalism, we do position ourselves as translators, we typically are more friendly to our sources than our counterparts in politics and business, and news stories that echo press releases are far more common than investigative stories. 

Many journalists on Twitter argued that we actually do need to have a conversation about why there isn’t more science journalism, and I agree. Francie Diep, who works at Pacific Standard and is freakin’ awesome, also made the excellent point that the article wasn’t primarily for science journalists; it was for people who read science news but might not know about these problems.

However,  to me, it sounded like the article’s author Michael Schulson wanted to blame the lack of investigative science journalism on self-declared science reporters not knowing that science is complicated and fraught with politics.

He writes, “At the same time, the conditions in which research takes place can incentivize scientists to cheat, to do sloppy research, or to exaggerate the significance of results. Simply put, science does have politics. There’s intense competition for funding, for faculty jobs, and for less tangible kinds of prestige” as if science reporters don’t know that.

We know that.

Schulson partially apologizes for his tone in that story in a follow-up post on Undark, but doubles down on the notion  that these political and economic issues are central to science (I agree), that his points have all been made before (Yes. Absolutely. Many times.), and that it would be irresponsible for science journalists to go around acting as if these problems in our coverage don’t exist. (Agreed.)

Here’s why the Pacific Standard piece still bothers me:

Presenting us science reporters as if we aren’t well aware that economics and politics shape–and often warp– science DISTRACTS from the conversation about why science news coverage is the way it is.

When I was reading the piece, I was seething due to feeling personally insulted rather than thinking about what I might be able to do to address the problem. That’s not productive.

I pitch a lot of news stories, because I’m a young science writer without a staff job. Pitching stories means selling stories, which means I have to convince editors that I can spin scientific studies into click-drawing narrative pieces. Science journalism’s tendency to focus on “Look at this cool new study!” type pieces is itself driven by the economic environment.

When you’re pitching, you don’t want to give editors reason to turn down the piece; they have enough already. Instead of disclosing your doubts and admitting that there were moments during the pre-pitch interview where you wondered if the scientist was lying, you weigh the evidence before hitting send.

When working up news pitches, most science reporters I know either: a) decide that the red flags are severe enough or numerous enough to justify scrapping the story idea

b) decide that the red flags are tenuous and present the story in its strongest possible light in order to maximize its chances of selling

Schulson probably knows that, but most people who aren’t writers don’t. I think we’d all be better off if people did know that, but nothing in the Pacific Standard piece would have helped a naive reader pick up on that dynamic.

(Furthermore, the two proposed solutions in the piece: 1) Lean on academics’ investigations more often and 2) Be more like John Bohannon , aka the guy who runs stings on scammy science journals and magazines,  are both non-starters.  There are many legitimate criticisms of Bohannon’s methods, not to mention the fact that we can’t all go around doing stings all the time. And if part of the problem is that we’re too dependent on scientific experts, how does collaborating with more experts on investigations solve the problem?)

I would love to be more able to discuss potential red flags that I see in studies with my editors, but there are a lot of constraints that restrict writers’ ability to do that, especially early in their careers.

Which brings me back to the tardigrades….

See a Red Flag; Say Something?

I cannot confirm or deny whether the plagiarism rumor I heard is accurate. There is, in fact, a third tardigrade genome sequencing project in Japan, but since the genetics institute didn’t post video or slides from the talk, there’s no way for me to track down the slides and check for plagiarism, at present.

(And arguably, I shouldn’t even be posting the rumor on my blog. Take the allegation with a grain of salt…And lime, because remember tequila was a factor in that conversation.)

The tardigrade genome controversy does stand out in my mind as an example of why it’s hard to write daily science news from an investigative angle, because even before the second tardigrade paper came out, there were a few signs that something might not be quite right with the first one and its extraordinary claims.

When I first read the PNAS paper, I noticed that a lot of its conclusions hinged on a very interesting but speculative explanation of how bacterial genes could have gotten into tardigrades.  Here’s how I described it in my write-up for MIT class:

“The paper’s lead author and a post-doc at UNC, hypothesized that picking up bacterial genes may be a side effect of being indestructible. Tardigrades’ famous resilience stems from their ability to shrivel themselves into tiny, dried-out kernels. Each can remain dormant in this form—called a “tun”—long enough to evade extreme temperatures and outlast fatal radiation. When they dry out their cells, their DNA becomes breakable, he argues.

When conditions improve, the tardigrades open up the pores in their cell membranes extra-wide to let water back in. Sometimes loose fragments of DNA from bacteria and other organisms might fall in through the pores, he says. Every once in a blue moon, a helpful bacterial gene might get stitched into a tardigrade chromosome and stay there for millions of years, the PNAS paper speculates.”

They did point to a couple of precedent papers, which showed that when dried plant seeds take water back in, plants do pick up stray genetic fragments. However, those plant papers were, in fact, part of the post doc’s graduate work. He used his own work on dehydrated plant seeds to back up his argument about gene scavenging tardigrades.

That struck me as suspicious early on in my reporting, but I had no idea what to do about it.  I mentioned that I had noticed the self-citing during the phone interview with the post doc but didn’t push on it.

I considered emailing one of the co-authors on the seed papers and asking whether that work really supported the tardigrade speculation. I should have but didn’t. November is one of the toughest months in the MIT Science Writing year, a time when you have to wrap up a 3000 word feature, produce a volley of news stories, write a chunk of your thesis, and still find time for your writerly lab observations, archival research, internship work, and dealing with whatever else is going on in your life. Even if I had gotten ahold of someone from the seed lab, I still would’ve need to interview a completely independent tardigrade expert. “Trust but verify” is a common mantra in science writing circles, but too often, when writers scramble to hit deadlines, “trust” ends up as code for “shortcut you use when TPTB have not provided time and/or resources to ‘verify'”.

Unfortunately, the scrambling doesn’t stop after grad school. Many staff writers are expected to produce a certain number of posts each week (or else they risk getting in trouble with their bosses), and freelancers have to write and pitch quickly enough to keep themselves fed.

The research needed to produce investigative-caliber work takes large amounts of time and energy, even on relatively short-and-sweet stories like the tardigrade genome story. It only took 9 days for a second team of researchers to put together a scientific critique of the paper. (To put that in perspective, when astronauts tested tardigrade’s ability to survive the vacuum of space, they left the tardigrades floating outside the International Space Station for 10 days. Tardigrades literally survived in the vacuum of space longer than this paper’s reputation survived Internet-ified biology.)

But in science journalism, nine days on a news piece is a luxury we don’t have. Even the science writers who know how to assemble genome reads would not have been able to empirically double check this finding without relying on the expertise of scientists.

Would I have been able to more carefully evaluate the paper’s claims with more time? Yes, absolutely.

But the pace of the science news cycle rarely leaves time for that level of paper deconstruction and analysis.

**tl;dr;**

We absolutely need more investigative journalism, and young science writers, especially, need more training on what to do when they spot a red flag in a study the editor wants to put in the daily news.

But it’s also important to acknowledge that the lack of investigative journalism isn’t due to science writers not caring or not knowing why they should investigate; it’s because of a lack of time, money, and training.

[Tune back in tomorrow for a follow-up blogpost with thoughts on what writers need to be able to make investigative science reporting happen.]

[EDIT: In response to criticism about my including a rumor in this blog post, I’ve edited the essay to remove the paper author’s name. I did my best to frame it as precisely what it is– an unreliable rumor– and I certainly did not intend to damage the reputation of the tardigrade paper’s author.  I am sorry for any embarrassment or complications my post may have caused.

At the same time, I think that removing the entire post from my site would be dishonest. My intent was to show through anecdote how difficult it is to assess a paper’s reliability on a science news schedule and how jarring it can be to find out (or start to wonder if) you missed something in reporting. If that didn’t come across in the post, that’s on me.]

Share on FacebookTweet about this on TwitterShare on Google+Share on LinkedInShare on TumblrShare on StumbleUponDigg thisEmail this to someone

2 thoughts on “We Need Investigative Science Journalism But Learning to Investigate Science is Hard-Part 1”

  1. How can you bring up a plagiarism rumor you heard from a tipsy person in a party without ANY evidence? Especially knowing that plagiarism is one of the worst accusations that can be made towards a scientist? Wow…

Leave a Reply

Your email address will not be published. Required fields are marked *

Website Protected by Spam Master