Good, Bad, or 'Necessary Evil'?

See allHide authors and affiliations

Science  30 Oct 1998:
Vol. 282, Issue 5390, pp. 860-867
DOI: 10.1126/science.282.5390.860

This special focus looks at the role of the embargo system in communicating scientific results to the public and to other scientists. Meetings and the special case of astronomy bring out some of the strains inherent in the system.

An arrangement aimed at keeping scientific findings out of the media until they are published by a journal draws mixed reviews; it is under pressure from Web-based publishing, and most physics publishers have already abandoned it

Every Wednesday or Thursday, more than 1400 reporters around the world get a sneak preview of the research articles that will appear in Nature a week later. The journal sends out faxes and e-mails highlighting the most newsworthy stories, and reporters can order the full text of any article. Two days later, more than 1200 journalists get similar advance notice of articles to be published in Science the following week. FedEx or priority mail brings early copies of medical journals like The New England Journal of Medicine (NEJM) and The Journal of the American Medical Association (JAMA). Reporters' e-mail inboxes and fax machines, meanwhile, fill up with announcements from other journals, universities, and institutes promoting new scientific findings. Most of this information carries a prominent warning: EMBARGOED. Public use of the information is forbidden until a specified date and hour to coincide with a journal's publication date.

What is most remarkable about this vast private traffic in science news is that it almost never leaks prematurely to the public. Hundreds of news-hungry reporters sit on the information, as they are bidden by journal publishers, until the designated release time. Welcome to the embargo system—a gentlemen's agreement between science journals and reporters designed to manage the flow of new scientific results to the public. The embargo system is the final stage of a process in which journals impose vows of secrecy not only on journalists but on the authors of the scientific papers they publish. No other area of journalism has such a cozy, formalized arrangement between reporters and their sources of news.

This odd system has developed and flourished over several decades because it offers advantages for everybody involved. Journals get maximum publicity, journalists get time to report complex stories, and scientists get more widespread and more accurate public exposure for their work. Indeed, the system is so successful that it has recently expanded with the debut of Internet-based clearinghouses that funnel embargoed information from a variety of sources to reporters who agree to abide by the rules. Behind the scenes, however, the embargo system is increasingly embattled.

It's a system wracked by built-in tensions. Science is supposed to progress through rapid communication of results among scientists, but the embargo system can erect barriers to this exchange of information. Nowhere is this more apparent than at scientific meetings, where scientists are often unclear on the rules for discussing results that are under review or in press at a journal (see p. 867). Newspapers and their reporters thrive on scoops, yet scoops are ruled out by the embargo system—and even some science reporters say the system encourages lazy reporting and undue attention to incremental advances. When a big science story comes along, however, competition is hard to suppress until a paper is published (see p. 862). Moreover, intense commercial interest in molecular biology has created new problems when information that can send a company's stock price soaring is distributed to hundreds of journalists under an embargo (see p. 865).

These built-in tensions are exacerbated by a new factor: the Internet and the World Wide Web. The Web is not only transforming scientific publishing, it's also changing the rules of the embargo system. In a world in which scientific papers can be disseminated to online subscribers as soon as they are accepted, the publication date of the printed version—and the embargo release time—becomes somewhat arbitrary. Moreover, the Web has created new avenues for circulating scientific information—from preprints of whole articles to bulletins of new astronomical observations—outside the embargo system, providing fodder for enterprising journalists (see p. 868).

All this is prompting many journals to rethink their embargo policies. Most physical science publishers have already abandoned the system, the American Chemical Society has virtually scrapped it, and even some biology and general science journals may follow suit. For example, Nicholas Cozzarelli, editor-in-chief of the Proceedings of the National Academy of Sciences (PNAS), says he's in favor of “getting rid of the embargo” in its present form and is proposing a new policy to his board this week.

This package of articles examines these issues from the perspectives of journal editors, reporters, and the scientists who are often caught in the middle. But first, by way of full disclosure, it should be noted that Science itself has a stake—or, rather, several different stakes—in the embargo system. The scholarly publishing side of the journal has a strict embargo policy (see Editorial, p. 877), and the American Association for the Advancement of Science, Science's publisher, has launched an ambitious Web-based clearinghouse for scientific information, EurekAlert!, that includes embargoed press releases. The News section, on the other hand, is on the receiving end of the embargo system: Science's journalists report independently on scientific developments published in this journal and others, and on data presented at meetings and elsewhere. Sometimes, the process even comes full circle when advance copies of Science news articles are distributed to other journalists under embargo.

Lofty purposes

Ask journal editors why they employ the embargo system, and the answer usually revolves around one issue: quality control. Insisting on secrecy from authors until their papers are published guards against public release of data that might not pass muster in peer review, and giving reporters a few days' advance access to papers that have passed review yields more accurate news. “The fundamental thing,” explains Jerome Kassirer, editor-in-chief of the NEJM, “is the protection of the peer-review process.” Says JAMA Editor George Lundberg: The system ensures that “quality is played out maximally in the public media.”

It was the NEJM that formalized the current system almost 30 years ago, when it published a set of principles known as the Ingelfinger rule, after the journal's editor at the time, Franz Ingelfinger. The Ingelfinger rule (see p.861) is still the guiding principle for the NEJM, but an estimated 300 other journals follow guidelines laid down by a group of medical editors calling themselves “the Vancouver group,” a reference to their first meeting place in Canada in 1978. The bottom line of their 25-page list of rules, updated most recently in 1997 (, is virtually the same as that of the Ingelfinger rule. Journals “do not wish to receive a paper on work that has already been reported in large part,” the Vancouver rules state, regardless of whether it has appeared “in print or in electronic media.” They warn authors to expect “prompt rejection” of any manuscript judged by editors to be a “duplicate publication.” Presenting the data at scientific meetings is fine, but sharing “tables and illustrations” with reporters is not.

The multidisciplinary journals have similar policies. Science uses “a variant of the Ingelfinger rule,” says Editor-in-Chief Floyd Bloom, “to educate the public broadly and accurately.” He says that there are benefits for scientists, too: Embargoes draw attention to new findings, and this builds public support for science. Publicity also attracts “the best authors.” Philip Campbell, editor of Nature, says his journal's embargo rules are motivated by a sense of “fairness”—a wish to make results available to “everyone at the same time”—and by a wish to maintain quality. But he also acknowledges some “self-interest,” in that the embargo system “maximizes the profile of the journal.” Publishers also argue that editing increases the value of articles and that the embargo system helps reward journals for their contribution. (Cell Editor Benjamin Lewin declined to discuss the embargo policy of his journal, which has taken a strong line on prepublication publicity, especially at meetings; see pp. 866 and 867.)

Medical editors cite another reason for embargoes: They don't want physician-subscribers to be caught off guard by stories in the media before they have the issue in their hands. Says Lundberg: “We believe that physicians have a right to have access to the full information in the article prior to being asked by patients to explain what the TV or the newspapers said about a drug they're taking or a disease they may have.” Richard Smith, editor of the British Medical Journal (BMJ) and a member of the Vancouver group, agrees, although he tries to take a flexible approach to embargoes. “It's in everybody's interest,” Smith believes, “to publish simultaneously the full scientific paper together with any media coverage.” That way, “if you're a critical reader, you can have a good stab at making up your own mind on whether you believe it or not.”

Once the journals are ready to publish, a multifaceted public relations enterprise swings into action, sending embargoed press releases from journals, institutions, and funding agencies to accredited reporters. Web-based science news services have recently sprung up to provide a central point for such information. EurekAlert! (, launched in May 1996 and financed in part by ads, is the prototype: It posts releases for university press offices, scientific societies, research institutes, publications, and government agencies in a public area and an embargoed news area, which 1860 certified reporters can access by password. Users pay nothing, but organizations pay up to $1000 a year to have material distributed. Adding to the PR blitz are several independent news services, notably Newswise (, which has scientific, medical, and academic clients similar to EurekAlert!'s; business services such as PR Newswire and the Dow Jones News Service; and an astronomy PR clearinghouse run by astronomer Stephen Maran.

Uneasy alliance

Journalists who use this embargoed news generally appreciate the ready access to privileged information and the extra time to prepare complex stories. Says Tim Friend of USA Today: “I don't support [the embargo system] for any deep moral or philosophical reason,” but “I do think it's useful. It gives us all time to do the reporting and research that's needed.” TV reporters are appreciative, too. “Embargoes are useful for us because TV has to get a picture to go with the story,” which takes time, says NBC science correspondent Robert Bazell. Bazell also likes the way embargoes create news, as it's hard to get on the air without an event. “We can all have broadcasts the night before [publication], run headlines the next morning, and it's news,” Bazell says. ABC's medical reporter Timothy Johnson sees the embargo system as an “honor code” among reporters that elevates the quality of information.

But journalists who benefit from the system are not dewy-eyed about its origins or its aims. Its chief purpose, many believe, is to generate publicity. “There's an awful lot of self-serving rhetoric about the orderly dissemination of information,” says Bazell, adding that it's “shot through with hypocrisy.” Dan Greenberg, founder and former editor of the biweekly Science and Government Report, allows “some rationality” in the idea that “you want to give science writers time to digest the material.” Moreover, he sees nothing wrong with seeking publicity, because “the first obligation of a publisher is to stay in business.” But he dismisses the high-minded defense of embargoes as wrapping “a selfish purpose in a flag of public good.” As for the argument that doctors need to get the news before their patients, it's “absolute nonsense,” according to Greenberg. “I don't think I've ever come across a physician who reads [NEJM] the instant it comes through the mail slot.”

Lawrence Altman, a science writer at The New York Times, may be the system's most dedicated critic. He speaks of the “greed” of the journals, which in his view purvey “taxpayer-financed research” and boost their prestige—and hence their circulation and ad revenues—with embargoed news releases. In a two-part essay in The Lancet in May 1996, Altman suggested that journals seek to “swell advertising coffers by intimidating scientists and physicians into silence.”

Others worry about the effects of the system on the way science is covered. Tom Siegfried, the science editor of the Dallas Morning News, says the system has “broken down from what it was intended to be”—a method of sifting wheat from chaff by helping reporters find the hottest news—and become “a barrier” to getting information. The worst effect is “what happens before a paper is submitted,” he says: Scientists won't talk about research they're developing for fear that publicity will kill the chances of publication. The result, Siegfried says, is that embargoes “prevent precisely the kind of reporting that most people think would be better”—the type that seeks to document the gradual development of knowledge. Instead, he sees embargoes contributing to hype about “breakthroughs.”

From the biological or biomedical scientists' perspective, however, the embargo may be a good thing, says molecular biologist Tom Cech of the University of Colorado, Boulder. It may chill relations with reporters a bit, Cech says, but “I think it inhibits people from making premature announcements” before their work has gone through peer review. That's just fine, he says, because “we shouldn't be rushing to the press.” Others are less enthusiastic. Neuroscientist Solomon Snyder of Johns Hopkins University believes it is mainly the “vanity of the journals” that sustains the embargo system. Nathaniel Landau, a molecular biologist at the Aaron Diamond AIDS Research Center in New York City, who canceled a public talk in 1996 to avoid jeopardizing a paper under review in Cell, says the Ingelfinger rule is really about selfpromotion. He questions whether journals “have any business” asking authors to be silent.

In spite of such complaints, most biology and medical journal editors—and the reporters who feed off them—seem to feel that the system's benefits outweigh its disadvantages, and they are prepared to hold the course. Says Lundberg: “I don't see [the embargo system] changing much in the near future.”

Cultural divide

Yet one substantial branch of scientific publishing has been undergoing a radical change of course: the physical science journals. Many journals in physics and astronomy once maintained strongly worded embargo policies, but they have gradually relaxed them in recent years. “It was certainly quite strict back in the good old days,” says Gene L. Wells, managing editor of Physical Review Letters (PRL), which has become the most prestigious journal in physics since its first issue on 1 July 1958. Now any restriction on publicity is at best informal, says Stanley G. Brown, administrative editor for The American Physical Society (APS), which publishes PRL and a number of journals focused on subfields of physics. Brown and Wells both say they doubt that early press coverage erodes the readership of their journals, pointing out that press reports seldom contain the scientific details of interest to readers of APS journals.

Embargo policies are no more draconian at the American Institute of Physics (AIP), an umbrella organization for APS and nine other learned societies, which publishes eight major journals itself, including Applied Physics Letters, Chaos, and Physics of Plasmas. Authors are simply asked—with little threat of enforcement—to wait until a paper is released to the printer before initiating any publicity, says Martin Burke, director of editorial operations at AIP. At that stage, peer review has run its course. Indeed, AIP itself often puts out an unembargoed tip sheet when a paper is accepted for publication in an APS or AIP journal, and reporters are free to write about the work well before it appears in print. Phillip Schewe, chief science writer at AIP, acknowledges that embargoes can catch attention: “There's nothing like putting an embargo on a press release to jack up the blood pressure of a reporter,” he says. But “it's pretty transparently self-serving.”

Topflight astronomy journals have followed the same route, relaxing previously strict embargo policies. “The change is that in recent years there have been huge numbers of reporters attending conferences,” says Helmut A. Abt, editor-in-chief of The Astrophysical Journal (Ap. J.), which is owned by the American Astronomical Society (AAS) and published by the University of Chicago Press. Reporters listen to talks or attend press conferences on results that will appear later in Ap. J. and Ap. J. Letters and write stories from the meeting, says Abt. “So we gave up trying to have an embargo,” he says. Paul Hodge, editor of The Astronomical Journal, another AAS publication, says that “when authors bring up the question about talking with reporters,” he asks them not to do so until a paper has been accepted for publication. But there is no sanction for not adopting the suggestion, and no paper has been rejected just because its content was publicized too soon, says Hodge.

Why the difference between the life and physical science disciplines? It could boil down to an ingrained openness that helped erode the embargo system from the inside, and the reality that few physics discoveries have an immediate impact on a company's stock price or a patient's questions, says Benjamin Bederson, a physicist at New York University who was editor of Physical Review A from 1978 to 1992 and editor-in-chief of APS from 1992 to 1997. “Physicists have not only been free in spreading their results—they're eager,” says Bederson. Asked whether there has been any change in the quality of press coverage of physics since embargoes have fallen by the wayside, Bederson says: “I didn't notice any serious change at all.”

The Internet: Changing the rules

Despite the wide-open attitude of physics publishers, many of them have long disliked one development: an electronic preprint server based at Los Alamos National Laboratory in New Mexico that freely distributes full-text copies of unpublished articles deposited there by authors. The archive ( is the work of physicist Paul Ginsparg, who began it in 1991. It signaled that the Web was about to change the rules of scientific publishing, providing a way to circulate papers widely outside the formal embargo system and potentially undermining conventional journals (Science, 9 February 1996, p. 767). That's exactly what Ginsparg intended. “Embargoes are clearly not in the best interests of scientists,” he said in an e-mail interview, adding that they “are shamelessly self-serving on the part of the journals.”

The archive posed an immediate challenge to journals that do not accept articles that have been published elsewhere. Most physical sciences journals have reluctantly decided, however, to consider papers that have been posted on Ginsparg's archive, although many would prefer not to. “It's a form of prepublication release,” says Alex Dalgarno, editor of Ap. J. Letters, “and it could impact the value of the journal.” The editors' dislike of the server is widely disregarded, says Frederick Lamb, an astrophysicist at the University of Illinois, Urbana-Champaign. Lamb says if journals decline to consider papers that have been posted on the Web, researchers would “vote with their feet … and just go elsewhere.”

Nature recently decided it will publish papers that have appeared on a public Web site. “Our policy,” says Editor Philip Campbell, “is that preprint servers are operating primarily as an intrascientific communication network and have the same sort of significance as a conference talk or list of published abstracts.” Internet release doesn't count as prior publication, he says, because the author is not implying that the article has been peer reviewed or that editors don't make an important contribution. “We haven't suffered yet,” Campbell says. Elsevier Science, adding yet another twist, says papers submitted to its journals may appear in a public archive or a home page as first drafts, but not editor-improved versions.

Science, however, is standing by its policy of not publishing papers that have been posted on the Web. Science Editor-in-Chief Bloom says: “If a paper has been publicly released on the Internet in the form that it was sent to us, then we consider that prior publication,” and Science may decline to take it. However, “if you assure us that you have a restricted site, we won't disqualify it” right off the bat. Monica Bradford, managing editor of Science, says physical scientists have been “very vocal” about their dislike of the policy. But, she says, “our physical sciences submissions have actually been on the increase, so I don't get the sense that it's been a problem.” She adds that rapid changes in the online world ensure that Science will continue to assess the policy.

So far, Web-based preprint publishing is mostly limited to the physical sciences. Ginsparg has opened a biology section in the archive, but entries are relatively sparse. And a separate venture run by HUM-MOLGEN, a nonprofit human genetics resource in the Netherlands, recently announced that it would post biology preprints after “low-key peer review” of submissions (Science, 19 June, p. 1807). But biologists are not yet clamoring to be published in it.

Nevertheless, at least one medical journal, the BMJ, is thinking the unthinkable: allowing potential authors to post electronic preprints on its own Web site. Editor Smith says BMJ already regards its Web site as the “primary” route of publication that has allowed it to reach “an entirely new audience” in the United States. His staff is now debating “whether to move to e-prints, as the physicists do.” BMJ might set up an area on its site where authors could post articles and receive comments, Smith says. If the author later wanted to submit the article for print publication, the BMJ would review it. “We're also looking at possibilities for doing peer review entirely openly on the Web,” says Smith: “I'm absolutely convinced that this is going to change everything.”

Few other editors are thinking of taking such radical steps, but a major scientific publisher, the American Chemical Society (ACS), has adopted a novel online publishing policy that changes the way papers are released to subscribers and the public. Beginning in January, the 26 ACS journals began releasing papers on the Web when they have been edited and checked by authors, sometimes as many as 11 weeks before they appear in print. ACS made the change because “authors wanted us to offer faster publication,” says publications director Robert Bovenschulte, adding that the decision was driven mainly by the technology. ACS felt it was embracing “the wave of the future,” adds ACS spokesperson Denise Graveline. Journalists are free to write about articles when they appear online, but this hasn't ended embargoes. Graveline says that ACS still notifies some journalists in advance of “a selected number of articles” before they are posted online.

Some medical journals have also used the Web for quick public release of papers that have important public health implications. Last year, for example, NEJM used the Mayo Clinic Web site to release a paper on heart valve injury associated with the fen-phen diet drug combination. And JAMA used the Internet last summer to distribute a paper on the adverse effects of a drug for hypertension. Lundberg says publishing online allowed the journal to post the full text, “bango, the same afternoon” that it cleared his desk. “Everybody responded beautifully, and we felt really good,” Lundberg says.

Does this new use of the Internet augur a major change in the way biology journals handle newsy reports? Lundberg is doubtful. High-priority articles are rare, he says, and JAMA is not planning to follow the ACS's lead yet and routinely post articles online before they appear in print. Kassirer, who says he tries “not to be too stiff-necked” about the rules, says things may change “over time … but at the moment, we are holding to our Ingelfinger rule.”

But some are ready to chuck tradition. PNAS's Cozzarelli, for example, would gladly go to early release on the Internet. “I believe that online preprints have made the embargo obsolete,” he says, and he'd like to rid science of the embargo system's “arbitrary” rules. But for many writers and editors struggling to keep up with science news, embargoes remain, as a biotech reporter says, a “necessary evil” that make the job more manageable.

Navigate This Article