From the issue dated January 10, 2003
|
|
|
Elsevier's Vanishing ActTo the dismay of scholars, the publishing giant quietly purges articles from its databaseBy ANDREA L. FOSTER Elsevier Science, it seems, is working on a disappearing act, and that has
Take the engineering article by Nikitas Assimakopoulos that was published in April 2000 in ISA Transactions, an Elsevier Science quarterly journal. A search in the company's ScienceDirect database for that piece, on work-flow management, brings up only the message, "For legal reasons this article has been removed by the publisher." Elsevier, the largest publisher of scientific journals, had concluded that the work was lifted nearly verbatim from a chapter of a book published in 1998 by two German professors. (Mr. Assimakopoulos, who is a professor at the University of Piraeus, in Greece, did not return e-mail messages or calls for comment.) So the publisher simply deleted the article from its online database. The Instrumentation, Systems, and Automation Society, which manages the journal, says it supports the decision. In the print journal, where removing an article is impractical, Elsevier published a retraction of sorts in the April 2002 edition. It noted that the article had been "previously published," and that "Professor Assimakopoulos has subsequently confirmed to the editor of ISA Transactions that this version of the article should be withdrawn from publication and that his name should not have been associated with it." The notice did not appear online. A recent online search indicates that the Anglo-Dutch publisher has quietly withdrawn dozens of journal articles from ScienceDirect since at least January 2000, often providing no details as to its reasons. In an e-mail message to librarians in November, Daviess Menefee, Elsevier's director of library relations for North and South America, wrote that the publisher had established a "strict internal policy" for the removal of articles, with the decisions being made by senior publishing managers and company lawyers: "Only under extraordinary circumstances can an article be removed (e.g., plagiarism, scientific misconduct, gross error such that human safety is at risk) and then a statement is inserted to indicate the removal." In an e-mail message, Eric Merkel-Sobotta, an Elsevier spokesman, says that the company "has not removed any article from ScienceDirect due to plagiarism." In an interview, he says the usual reason for expunging articles from ScienceDirect is fear that Elsevier will be accused of copyright infringement by a scholarly society or author. The publisher, he says, wants to avoid a judge's order requiring the removal of certain articles. (Elsevier Science has been accused of copyright infringement in at least one lawsuit. Mr. Merkel-Sobotta would not say whether the publisher has been ordered by a court to remove an article from its database.) The reason for providing no details beyond the bare statement -- "For legal reasons this article has been removed by the publisher" -- is that to offer more specifics would be tantamount to admitting liability, says Mr. Merkel-Sobotta. He says Elsevier has expunged no more than 30 journal articles from ScienceDirect. That total represents only a fraction of the millions of articles in Elsevier's database. But that is small solace to librarians and scholars who study the history of science and medicine. As more and more scientific literature is moved from print to a digital environment, they fear, holes in databases could leave researchers ignorant of why certain articles were considered questionable -- and may even lead to poor medical treatments and faulty scientific research. And, scholars ask, shouldn't researchers be warned about authors who plagiarize, or commit scientific fraud or misconduct? "I don't think people should take scientific material and make it disappear, especially if it's gone through peer review," says Mark S. Frankel, director of the program on scientific freedom, responsibility, and law at the American Association for the Advancement of Science. "There should be a digital trail which allows these things to be seen, observed, and studied." Dangerous Precedent Though Elsevier deletes more articles than anyone else, it is not alone in swinging the ax. LexisNexis and Westlaw, two major providers of online legal texts, removed an article from their databases after its integrity had been questioned. The article, written by two business professors and published in the spring 1998 issue of the Denver Journal of International Law and Policy, criticized multinational corporations, particularly Boise Cascade. Critics of the piece quibbled with some of its reporting, but there was no proof of plagiarism or other publishing malfeasance. Even so, after the company reacted harshly, the University of Denver, which publishes the journal, successfully pressured LexisNexis and Westlaw to pull the article from their databases. The authors settled their lawsuit against the university in 2001 (The Chronicle, September 7, 2001). Under the agreement, the university apologized to the authors and paid them an undisclosed amount. The most notorious expunging of a scientific journal involves Elsevier's removal of an article published in September 2001 in Human Immunology. The paper, about the genetic origins of Palestinians, generated a political firestorm because it labeled Jews living in the Gaza Strip as "colonists" and said some Palestinians were living in "concentration camps" (The Chronicle, November 23, 2001). Elsevier Science removed the electronic version of the article and sent a letter to subscribers telling them to ignore it in the print edition or, preferably, to "physically remove the relevant pages." The letter said the article "included certain statements" that the journal's owner, the American Society of Histocompatibility and Immunogenetics; the editor in chief; and the company "found were entirely inappropriate for articles published in this journal." The former president of the society said the article offered opinions about the politics and history of the Middle East when it was meant to be only a science article. The incident sent shock waves through the world of health-science librarians, and put librarians on notice that Elsevier's business included unilaterally stripping from its databases any articles that it deemed flawed or otherwise unworthy. Impact on Scholarship T. Scott Plutchak, director of the health-sciences library at the University of Alabama at Birmingham, warns that the company's actions could have severe repercussions for history and scholarship. "We must never forget that the preservation of the historical record, with all of its faults, mistakes, and corrections, is an essential part of the service that librarianship performs for society," he wrote in an April 2002 article in the Journal of the Medical Library Association. "As the medium of information becomes more elusive, we must become more vigilant." A continued fragmenting of databases could eventually hamper research into the history of science, says Bruce V. Lewenstein, an associate professor of science communication at Cornell University. "We're not yet, for the most part, at the point where we're tracking down articles electronically when we're studying the history of a particular field," he says. "But we soon will be." Elsevier says it respects the historical record, and plans to archive all of its electronically published journal articles with the Royal Dutch Library, in The Hague. The expunged articles are not currently available online through the archive, which is being developed. Whether the articles will be electronically accessible through the archive is under discussion, says Karen Hunter, senior vice president for strategy at Elsevier Science. But part of the historical record, some scholars argue, is highlighting missteps and unethical behavior. If a journal article proves to have been plagiarized or to reveal scientific fraud, they say, other researchers should know to be wary of the authors. And they argue that even if Elsevier decides to make the deleted articles available through the library, an overseas database that is not readily available to American scientists will do little to forewarn them of the tarnished authors' behavior. Even R. Russell Rhinehart, editor in chief of ISA Transactions -- which published Mr. Assimakopoulos's electronically withdrawn piece -- says authors who have plagiarized should be "publicly embarrassed." Their articles should be kept in the database and permanently stamped with an erratum, says Mr. Rhinehart, who is head of the School of Chemical Engineering at Oklahoma State University. "Everybody knows you don't plagiarize. You don't take somebody else's idea and claim it's your own." Some say there are dangers associated with expunging material that goes beyond stymied scholarship or the cloaking of plagiarists. Drummond Rennie, a deputy editor at The Journal of the American Medical Association who is a professor of medicine at the University of California at San Francisco, says pulling a fraudulent article from a database could lead to poor treatment for patients. Physicians, unaware that the article was yanked, might remember reading it and act upon it. For example, he says, a physician might prescribe medication based on an article that incorrectly states the drug's effects. What's more, if an expunged article is not tied to a notice explaining why it has been retracted, researchers may never learn that the article they once read or heard about -- even though it has since been removed from electronic circulation -- has been discredited. "We know that if you don't make this tight binding of the retraction to the paper, then people will go on citing the paper," says Dr. Rennie. He wrote an article in JAMA in 1994 that described how physicians continued to reference an article by the cardiologist Robert A. Slutsky long after scholars had determined that it was fraudulent. To such criticisms, Mr. Merkel-Sobotta, the Elsevier spokesman, responds that the publisher does not "undertake lightly" the withdrawal of articles from ScienceDirect. It does so because of legal concerns, and only on the recommendation of its "board of article withdrawal," led by Elsevier's director of academic relations. Publishers are free to establish any policy regarding retractions of articles from online databases. "It is entirely up to the publishers what they do with their online information," says Harriet de Hoog, a spokeswoman for the International Association of Scientific Medical and Technical Publishers. But five science publishers interviewed for this article say they do not remove plagiarized or otherwise defective articles from databases -- and these publishers suggest that Elsevier's policy is a dangerous anomaly. The usual practice is to maintain the original article and attach a correction that explains why the article is in error. In that way, a search for the defective article brings up the erratum or retraction notice. Such a standard, says Mr. Frankel, of the American Association for the Advancement of Science, ensures that researchers aren't confused or duped, and that they don't waste time online trying to track down retracted articles. JAMA has never found an article so faulty that it considered a retraction, says Margaret A. Winker, a deputy editor, who directs the journal's division of scientific online resources. But if it did, she says, it probably would follow a procedure similar to the one it uses in making corrections to electronic versions of articles. A retraction would be linked to the original article, which would also have the retraction appended to it. Similarly, the journal Science does not remove electronic versions of inaccurate or plagiarized articles, says the executive editor, Monica M. Bradford, but leaves the original article online and links it to a retraction notice. For example, Science did not remove the electronic versions of several articles it published in 2000 and 2001 by Jan Hendrik Schön, which investigators later determined were based on phony experimental results (The Chronicle, June 7, 2002). Instead, the journal added retraction notices at the bottom of each online article. In addition, to the left of each article, a "Related commentary" box contains links to the retractions at the bottom of the articles. The policy is a work in progress, Ms. Bradford acknowledges. "It is our intention to get the word 'retraction' more prominently on the articles." The American Physical Society did not remove electronic versions of two of Mr. Schön's articles from Physical Review Letters, instead linking the articles to the errata. The articles were also linked to a report from Bell Labs, which investigated charges of misconduct against the author, says Martin Blume, editor in chief of the society. "I've called it an electronic pillory," he says of the society's decision to keep faulty articles online and connect them to correction notices. A Different Approach A slightly different approach was taken by the Institute of Physics, which decided not to withdraw a widely criticized paper published in November 2001 in Classical and Quantum Gravity by the French twin brothers Igor and Grichka Bogdanov (The Chronicle, November 15, 2002.) Many scholars said the article, "Topological field theory of the initial singularity of spacetime," was nonsensical. The brothers maintain their work is valuable and important, and say the critics hold grudges against French scientists and members of the French publishing industry. "At present, there are no plans to withdraw the article. Rather, the journal publishes refereed 'Comments and Replies' by readers and authors as a means to comment on and correct mistakes in published material," wrote Andrew Wray, senior publisher of the bimonthly journal, in an e-mail message. The American Geophysical Union does not remove discredited journal articles from its electronic database, says Judy C. Holoviak, deputy executive director of the society. "Once you do that, the scientific community will not know what other thing you might have done to the record," she explains. Concern over preserving the historical record is prompting some scholars and librarians to urge publishers and editors to develop standards for the electronic handling of inaccurate journal articles -- standards that do not include their deletion from online databases. That expungement, says Mr. Lewenstein, will cause "practical problems for both active researchers and then long into the future, when historians of science are trying to understand the history of a particular field." ARTICLES THAT HAVE VANISHED Elsevier Science purged these articles from its ScienceDirect database. Article: Workflow management with systems approach: anticipated and ad-hoc workflow for scientific applications, by Nikitas Assimakopoulos. Journal: ISA Transactions, April 2000 ScienceDirect notice: "For legal reasons this article has been removed by the publisher." Reason for withdrawal: Erratum in April 2002 of ISA Transactions says article had been lifted from a book chapter. Articles: Epidermal growth factor receptor: a transcription factor? and EGFR as a transcription factor? by Richard W.C. Wong and Siu-Yuen Chan. Journals: Trends in Biochemical Science, Nov. 2001; Trends in Genetics, Nov. 2001; and Trends in Endocrinology and Metabolism, Dec. 2001 ScienceDirect notice: "For legal reasons this article has been removed by the publisher." Reason for withdrawal: Apologies in the three journals say large parts of the articles had been copied from a previously published Nature Cell Biology article. Article: The Origin of Palestinians and Their Genetic Relatedness With Other Mediterranean Populations, Antonio Arnaiz-Villena, lead author. Journal: Human Immunology, Sept. 2001 ScienceDirect notice: None Reason for withdrawal: Articles dealt with political and historical, rather than scientific, issues.
SOURCE: Elsevier Science, Chronicle reporting
http://chronicle.com Section: Information Technology Volume 49, Issue 18, Page A27
|
Copyright © 2003 by The Chronicle of Higher Education