Why I hate science by press release

I read today about another exoplanet that is high up on the Earth Similarity Index, meaning it has a reasonable likelihood that it is habitable. By habitable, we normally mean that it’s at the right distance from the star that liquid water might exist on the surface, NOT that it has life on it, or even an atmosphere. The planet in question is Gliese 163c, which comes in at number five on the list of known planets in the Habitable Exoplanets Catalog. This is exciting stuff in an exciting – and cutthroat – field of astronomy and astrophysics.

I was reading about the system in an article on Discovery News: it’s made up of two planets, one a so-called “Super-Earth” in the habitable zone of the cool red star Gliese 163 (a Super-Earth is something that’s Earth-like but bigger). It was found by a European team using a purpose-built spectrograph to look for planets. They’ve found a lot of the planets that we’ve discovered from the ground. All good so far. Then I got to this line:

The research has been submitted for review and publication.

Wait, WHAT? You mean it HASN’T been peer-reviewed yet? They put out a press release about an exoplanet discovery BEFORE anyone else had independently verified the discovery or check their methodology and interpretation? That is dangerous, and to be frank, downright irresponsible.

Okay, I said I’d try to keep rants to a minimum in this blog. However this is an issue that has reared it’s ugly head too many times before in science. Perhaps the most famous example of this “science by press release” is the Fleischmann-Pons Cold Fusion debacle. In that case, the “results” generated a lot of media attention, due to their implications, but was soon discredited to the point where the phrase “Cold Fusion” sends researchers into a cold sweat and is avoided wherever possible.

I should point out that I’m not saying that the Gliese 163c result is wrong. But one of the KEY things about science is the peer-review process. If you subvert it in this way (by going through the media first), and then get it wrong, you end up with egg on your face, and the general reputation of scientists takes a hit. And the thing is, this team has a history on this front.

In 2007, they announced that the first detection of an exoplanet in the habitable zone of any star. The planet, a Super-Earth called Gliese 581c, had a mass of 5 times that of Earth and took only 15 days to orbit its Gliese 581, but since the star is a cool red dwarf, the team claimed that it was in the habitable zone. They had a done a back-of-the-envelope calculation to come to that conclusion.

This result garnered a huge amount of media attention. The planet was referred to as “Earth II” in headlines around the world. I remember commenting on the result to the media at the time, and warning that the paper had only been submitted when the press release went out. This is a perhaps subtle point that not many seemed to pick up on.

After a day or so, the submitted paper was made available to the world, and other astronomers did some more detailed calculations that showed that the planet was not in the habitable zone at all! Another planet in the three-planet system is in the habitable zone, but it weighs in at almost 8 times the mass of Earth, so it is far less likely to be habitable. It is this planet that quietly became the poster child of the Search For Life. But what if the third planet around Gliese 581 wasn’t there?

So – the announced result was wrong, but kind of right in the end, so why the fuss? Because this is not how science works! What’s the point of the peer-review process if you announce your results to the world first? The general public and media will simply accept the result as correct and move on (Cold Fusion was an exception case). This kind of thing leads to sloppy science, which in turn can lead to a mistrust in scientists. The last thing we need at the moment!

I note that the Higgs Boson result was accepted recently by Physics Letters B as two papers, three months after it was announced to the world, so this phenomenon is not limited to astronomy and astrophysics.

Has science now reached the stage where it is really better to be first than right? It would seem so.

Aside: The Gliese 581 planetary system is controversial, but not because of planet c or d. An independent team claimed the discovery of another planet that – if confirmed – would at the top of Earth Similarity Index as the most habitable planet we know of. Trouble is, a lot of people don’t believe the result. That’s story for another post though.
Advertisements

About iwouldntnormallycall

I'm an astronomer and science communicator, these are my adventures. Views posted are my own.
This entry was posted in Astronomy, Outreach, Research, Science and tagged , , . Bookmark the permalink.

5 Responses to Why I hate science by press release

  1. Mags says:

    Well said.

  2. marksackler says:

    According to the “scoreboard” on the JPL Planetquest site, there are 809 confirmed exo-planets and 2,321 candidates. This new announcement would seem, at best, to be a new candidate. It might turn out to be not even that, it might be a candidate that has already been put in the queue for confirmation by a more responsible source that did not announce it yet. Remember cold fusion?

  3. Kelly says:

    Announcing a new exoplanet before a paper’s been through review is, of course, not the best idea, but that doesn’t mean that everyone will just take the press release at its word. A good science journalist will show a paper to several independent scientists to get a feel for any issues before they report on it — putting it through an informal peer review, essentially. So it’s only the less scrupulous members of the press you need to worry about hyping something like this.

    Saying that I don’t see why a scientist would want to risk publicising their results before they’re sure of them, and peer review would certainly give them an extra layer of protection. Fear of being scooped, maybe?

    Also worth noting that papers coming from CERN, like the Higgs one, are a bit different because they have so many internal layers of review before anything even gets on to arxiv.

    • I think you raise some good points Kelly.

      I agree that a good science journalist will get other scientists to look at the work, although sometimes all you have is a press release. I’ve been asked to comment on press releases without having the paper to hand before. Tricky to do! The other problem is that in some situations non-science journalists will be covering the story (not every media outlet has a science reporter/journalist anymore, sadly). In the case of Gliese 581c, I think it was the media that called it “Earth II”, not the scientists or their press office.

      I think the fear of being scooped is a big reason for science by press release. There have been tussles over who discovered what planet first (for example). When the story is possible life-bearing planets, I think everyone wants to be first. This is big news: Marcy and Butler were on the front cover of Time magazine for example!

      On the internal layers of review, I agree that helps; it’s one of the reasons I mentioned the Higgs Boson only in passing. However, to me, independent review is still more important. As collaborations and teams get larger and larger, proper independent review is getting harder to do, since almost everyone who knows anything about the project is involved in some way. I think that’s something that scientists and journals will have to address in the coming years.

  4. Pingback: Science spinning out of control? A planet by any other name | I Wouldn't Normally Call

Comments, Queries, Suggestions?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s