The revolutionaries were too early.

The movement to make biology papers freely available before they have been peer-reviewed, let alone published in a reputable journal, finally succeeded in 2013, when bioRxiv (pronounced bio-archive) was launched by Cold Spring Harbor Laboratory. But 50 years before, the National Institutes of Health tried something similar: distributing unpublished scientific papers, or preprints, to a handpicked group of leading researchers.

The effort was intended to speed the dissemination of potentially important advances, but it was met with such hostility from some eminent biologists and journals — one called the papers “shoddy merchandise” — that it was shut down after just six years, a historian reported on Thursday in PLOS Biology.


“It’s fascinating to see that the same things happened 50 years ago,” said bioRxiv co-founder Richard Sever, who had never heard of the earlier preprint effort (which, of course, used snail mail). “The business concerns of the journals, the scientists who warned about the terrible things that would happen if information that wasn’t peer-reviewed got out — that was very much what we experienced with bioRxiv.”

The long-forgotten NIH effort began in 1961, when an official named Errett Albritton, then 70, dreamed up “Information Exchange Groups.” They initially consisted of “leading investigators” in a narrowly defined field, such as “computer simulation of biological systems,” found Matthew Cobb, a scientist and historian at the University of Manchester, who stumbled on dusty documents about it in the archives of Cold Spring Harbor Lab.

Researchers sent in drafts of their soon-to-be-published (they hoped) papers, and the NIH made copies and mailed them to members of the relevant specialty group. The idea, as one supporter wrote, was to allow scientists “to be fully informed in record time of all important developments in the field.”

Not everyone saw it that way. Francis Crick, co-discoverer of the double helix structure of DNA, told Albritton in a 1961 letter that “there is far too much careless and rapid communication already in every area of this field of study,” referring to genetics. “The idea of increasing it even in this semi-public manner fills me with horror.”

When Cobb stumbled on the letter from Crick in the Cold Spring archives, “it wasn’t clear to me exactly what scheme Albritton had proposed; I had never heard of him,” Cobb said in an interview. “So I dug out the letter that Albritton had written, and it was clearly quite interesting. I then just pulled on the thread … and gradually uncovered the whole story. Although the IEGs were well known in the 1960s, and have been mentioned a couple of times by historians of information, scientists have completely forgotten this story.”

Crick’s reaction was mild compared to the fury of leading journals and scientific societies, many of which publish their own highly lucrative journals. The editor of Science, published by the American Association for the Advancement of Science, wrote in a 1966 editorial that the preprints the NIH was circulating were “government-subsidised shoddy merchandise,” revealing “a desire on the part of some scientists to avoid a discipline essential to the integrity of science” — namely, waiting for peer review and acceptance in a journal before informing the world of their experimental results.

The American Association of Immunologists, publisher of The Journal of Immunology, called the NIH’s actions “improper” and fretted that the preprints (which the NIH called memos) might “ultimately supersede” published papers. Any research circulated by the NIH program would be banned from publication in its journal, the association decreed. Nature, then as now one of the world’s leading journals, saw circulating preprints as “next to downright villainy,” and said “misguided zeal is one of the most dangerous forces in society.”

“The publishing establishment saw this is a major threat to both their prestige as gatekeepers and, in particular for the learned societies, who then as now make a substantial part of their income from publishing, a threat to their finances,” Cobb said.

The death knell for the preprint project came in September 1966, when other journals followed the immunologists’ lead. They would not accept any paper that had been previously circulated. Faced with being barred from publication — the coin of the realm in academic science — researchers who had shared unpublished work gave up. After circulating some 2,500 preprints, the NIH ended the program in March 1967.

Journals kept biology preprints from being shared before publication for decades after. A physics preprint server, arXiv, was launched in 1991, however, and when the world did not end (and physics journals didn’t disappear), biologists tried again. A 1999 effort by the NIH was met with attacks like those in the 1960s, however. The New England Journal of Medicine called it “a potential threat to the evaluation and orderly dissemination of new clinical studies,” and the Federation of American Societies of Experimental Biology threatened to lobby Congress to cut the NIH budget if the proposal went ahead.

It didn’t. But biologists in favor of preprints finally succeeded with bioRxiv.

One reason, Sever said, is that he and others at bioRxiv persuaded leading journals that posting a preprint was little different than giving a talk at a science meeting, which journals had not deemed a disqualification from publication. Hundreds of journals now accept papers that were posted months or even years earlier at bioRxiv.

One reason preprints have become accepted is the recognition that they “speed up research,” Sever said. “Those months or years between when a paper is submitted to a journal and when it’s published is time when people in the field can be looking at your data and building on your results.”

The preprint system has worked so well in basic biomedical research that bioRxiv plans to launch a preprint site for clinical trials and epidemiology, Sever said. It will be a partnership between Cold Spring Harbor Lab and Yale University, he said, and should go live in 2018.

Leave a Comment

Please enter your name.
Please enter a comment.

  • >The preprint system has worked so well in basic biomedical research that bioRxiv plans to launch a preprint site for clinical trials and epidemiology, Sever said. It will be a partership between Cold Spring Harbor Lab and Yale University, he said, and should go live in 2018.

    The prerequisite to making this a meaningful effort will be to get clinical trial investigators and funders to report their results at all.

  • Have you overlooked and Many reports on those sites meet any definition of “preprint.” It is common for authors to submit descriptions of work in progress seeking comments that will help them surmount problems impeding progress. FDN

A roundup of STAT’s top stories of the day in science and medicine

Privacy Policy