Reporting on Science: Does the Press Get it Right?


“We don’t serve faster-than-light particles here,” growled the bartender. A neutrino walks into a bar.

Last week, scientists at the CNGS experiment (CERN Neutrinos to Gran Sasso) reported the arrival in a lab in Gran Sasso, Italy, of neutrinos produced at the accelerator in CERN, on the Swiss-France border, at a rate that implied they were moving slightly faster than the speed of light. As soon as the reports hit the press, in physics departments around the world, jokes like the one above were all the rage. Particles moving faster than light? Wouldn’t that mean a violation of causality? Could these particles be moving backwards in time?

Behind the science is an interesting social issue, however… how much can you believe of you read in the papers about science? Do news reports of major breakthroughs get it right?

The role of scientist as newsmaker has a long history, but perhaps a key moment occurred in 1919, when Sir Arthur Eddington announced to the Royal Society in London his observation of an apparent shift in the position of a star seen near the Sun during a total eclipse, which he pointed out was evidence in favor of Einstein’s General Theory of Relativity.

The Times of London carried the story under the headline “Revolution in Science: New Theory of the Universe, Newtonian Theories Overthrown.” This caught the attention of the New York Times, who asked one of their reporters in London at that time to follow up on the story. 

Their correspondent, Henry Crouch, was not an expert in science; he was a sports writer in England reporting on golf. As described in Walter Isaacson’s recent book, Einstein: His LIfe and Universe, Crouch couldn’t get into Eddington’s conference at the Royal Society so he telephoned Eddington himself for a comment. On November 9th, his article appear on page 6 of the Times under the headlines “Eclipse Showed Gravity Variation: Diversion of Light Rays Accepted as Affecting Newton’s Principles. Hailed as Epochmaking.” Crouch opened his article by attributing to Sir Joseph Thomson, President of the Royal Society, the comment that the discovery was “one of the greatest — perhaps the greatest — of achievements in the history of human thought.” Of course, Crouch wasn’t actually there to hear Thomson’s remarks.

The next day, this story was followed up (on page 17) with the far more entertaining headlines: “Light All Askew in the Heavens / Men of Science More or Less Agog Over Results of Eclipse Observations / Einstein Theory Triumphs / Stars Not Where They Seemed or Were Calculated to be, but Nobody Need Worry / A book for 12 Wise Men / No More in All the World Could Comprehend It, Said Einstein When His Daring Publishers Accepted It.”

As Isaacson wryly comments, it was “a classic from the days when newspapers knew how to write classic headlines.” Meanwhile, Crouch’s bewilderment shines through right in his lede: “Efforts made to put into words intelligible to the non-scientific public the Einstein theory of light proved by the eclipse expedition so far have not been very successful…” And indeed, the “only twelve persons in the world can understand Relativity” canard got its start at the end of this second article.

The story of this story illustrates the constraints that science journalists work under. A journalist is rarely a specialist in any news story, and must do his or her best to grasp the details well enough to make them comprehensible to every reader — then move on to the next story and do the same thing again. Banking, baking, bolometry, a reporter is frequently a jack-of-all-trades. In Crouch’s case, it is understandable that a golf writer would report a scientific news conference in the only way that he was used to writing.

A second issue, of course, is that even the best reporter is at the mercy of his or her editor, not to mention the headline-writer. 

But a third problem is in the nature of news reporting itself. News must be significant, but it must also be… well… new. And usually scientific advances are not recognized in a fashion that fits a typical reporter’s deadline. Thus easy-to-identify moments, like a press conference, are the hook for stories whose origins may actually have arisen years earlier. After all, Einstein’s General Relativity was actually published three years before this announcement, in 1916. And indeed, it has been argued that Eddington’s confirmation was premature — the error in his data were so large that by themselves they did not actually confirm Relativity.

But in fact the biggest problem of science reporting is not with the reporters, but with the scientists themselves. Big Science requires Big Bucks (“No Bucks — No Buck Rogers” as Tom Wolfe quotes Gus Grissom saying, in The Right Stuff). Science requires public support, which in turn means public awareness. The Hubble Space Telescope is famous not only for its good science but also because it’s got a first-rate publicity department.

When does Big Science decide to go to the papers? Sometimes you are really sure you’ve got something, even if others are still skeptical; then you call a press conference — usually after the paper has been accepted at a peer-reviewed journal. This was the case with the faster-than-light neutrinos. Sometimes word leaks out, and you are rushed. This happened in 1996 with the Mars meteorite that was thought to show evidence of fossil life forms.

But sometimes the announcements are premature. For example, CERN’s UA1 collaboration announced they had the top quark in 1984; it turned out, they didn’t. (The top quark was finally found in 1995 by experimenters at Fermilab.)

Brad Schaefer, an astronomer then at Yale, in 1999 took more than 400 news articles about gamma-ray astronomy, supernovae, and Mars and passed them out to a number of astronomers (including one of us, GJC) to look for errors, ranging from the fundamental to the trivial. The articles came from a number of general news sources, ranging from UPI and the New York Times to specialists like Sky and Telescope and Science News, covering the years from 1987-1996.

Not surprisingly, the specialist publications almost never made trivial errors, while the various newspapers averaged roughly one trivial error every two articles. 

When it came to more serious mistakes, though, he found that in fact “none of the articles significantly mislead the reader or misrepresented the science.” But when the astronomers rated each story with the probability that its basic new science claim was correct — in many cases, judging with hindsight not available to the original reporters — they found that the average probability of accuracy over all stories was only 70%, “regardless of source, topic, importance, or quoted pundit.” 

The errors came not from the reporting, but from the scientists themselves. Only about 70% of the time do scientists’ newsworthy announcements hold up over time. Schaefer points out that you need to make a distinction between textbook science (with reliability near 100%) and the science that you read in the newspapers.

Journalists are only as reliable as their sources. But given that caveat, they actually do a pretty good job.

And it’s an important job. Science reporting is the first way that we learn about new discoveries. Writers must get new science ideas somewhere, before they can become turned into science fiction. Once that happens, these ideas can enter the popular imagination. And the ultimate sign of a successful new idea, of course, is when it can become the next good joke.

How can you spot neutrinos on the Enterprise? They’re the ones wearing red shirts at the “end” of the episode…

Brother Guy Consolmagno SJ is a Jesuit brother and planetary scientist, the curator of meteorites at the Vatican Observatory, with a lab in the gardens of the Pope’s summer home. He’s written several books on astronomy, and on science and religion; his latest, with Dan Davis, is a new edition of the classic handbook for using a small telescope, Turn Left at Orion.

William S. Higgins writes and speaks about science, technology, and history. He is a radiation safety physicist at Fermi National Accelerator Laboratory in Illinois.


Subscribe to this thread

Post a Comment

All comments must meet the community standards outlined in's Moderation Policy or be subject to moderation. Thank you for keeping the discussion, and our community, civil and respectful.

Hate the CAPTCHA? members can edit comments, skip the preview, and never have to prove they're not robots. Join now!