Some Research Wants to Be Free, Some Follows the Money: Bogus Journals Complicate the Open Access Movement

Abstract

A glut of junk open access journals, pandering to desperate authors driven by the “publish or perish” mantra, is staining the reputation of that new model in scientific and scholarly publishing.Despite the appeal of Stewart Brand's rallying cry “information wants to be free,”1Clarke R. “Information wants to be free.”. Xamax Consultancy/Australian National University, 1994−2009http://www.rogerclarke.com/II/IWtbF.htmlGoogle Scholar the ambiguities of that key term free have long fostered confusion. Does information gravitate toward freedom in the sense of free beer or a free lunch, raising the obvious objection that there is no such thing? The costs of generating and distributing information must be recovered somewhere. Or does freedom in the sense of free speech, unfettered by official censorship, steep access costs, or other artificial constraints, better suit the purposes of scientific communication?Digital distribution has reduced economic bottlenecks in the circulation of research findings and allowed the creation of various structures that offer open access (OA) publication. The Public Library of Science (PLoS), offering 7 peer-reviewed and professionally respected journals (including PLoS One, which has become active in emergency medical research2Greenberg C.J. PLOS ONE: leading emergency medicine journal?.http://openbiomed.info/2012/11/plos-em-journal/Google Scholar), is the best known of the OA entities that have arisen since World Wide Web access reached the general public in the early 1990s. Others include Biomed Central; the British Medical Journal (for research articles, not full content) and its BMJ Open offshoot; Hindawi, a large Cairo-based publisher that has courted controversy but meets most experts' credibility criteria, and that converted all its journals to OA by 2007; and Paul Ginsparg's arXiv.org, which has made self-archiving the norm in physics and commonplace in related disciplines. Supported by assorted mechanisms, including author fees, institutional subsidies, and voluntary or mandatory pay wall–free archiving, OA publishing represents a dramatic step toward an environment in which research information resembles Peter Cooper's ideal of education as a public good, “as free as air or water,” rather than a commodity that can be bought, sold, and restricted to those with the ability to pay.Perhaps inevitably, note observers of the OA movement, enterprises with a sharper eye for authors' wallets than for readers' interests have launched journals that accept practically any article for a fee, perform little or no peer review, and threaten to dilute the credibility of the OA scientific literature, at least in the eyes of the unwary. These on going concerns play on the unfamiliarity of many journals outside specialized communities and the economies of scale inherent in online publishing. The pay-to-publish operations are numerous enough—and enough of a nuisance to researchers, tenure and promotion committees, and scientific communication specialists—to have provoked a watchdog response.3Butler D. Investigating journals: the dark side of publishing.Nature. 2013; 495 (Accessed May 27, 2013): 433-435https://doi.org/10.1038/495433ahttp://www.nature.com/news/investigating-journals-the-dark-side-of-publishing-1.12666Crossref PubMed Scopus (236) Google Scholar The flood tide of suspect publications and related conferences, amounting to “a parallel world of pseudo-academia,” came to the general public's attention last spring through a New York Times article.4Kolata G. Scientific articles accepted (personal checks, too).New York Times. April 8, 2013; (Accessed June 2, 2013): A1http://www.nytimes.com/2013/04/08/health/for-scientists-an-exploding-world-of-pseudo-academia.html?ref=health&pagewanted=allGoogle ScholarThe openness of the digital realm attracts both high- and low-quality material; as standards evolve, it is anyone's guess which category will dominate online scientific reporting. OA has the potential to expand and transform scholarly publishing, its proponents contend,5Harnad S. Scholarly skywriting and the prepublication continuum of scientific inquiry.Current Contents. 1991; 45 (Accessed May 29, 2013): 9-13http://cogprints.org/1581/1/harnad90.skywriting.htmlGoogle Scholar, 6Harnad S. Post-Gutenberg galaxy: the fourth revolution in the means of production of knowledge.Public Access Comput Syst Rev. 1991; 2 (Accessed June 3, 2013): 39-53http://cogprints.org/1580/1/harnad91.postgutenberg.htmlGoogle Scholar, 7Harnad S. Ethics of open access to biomedical research: just a special case of ethics of open access to research.Philos Ethics Humanit Med. 2007; 7 (Accessed June 4, 2013): 31https://doi.org/10.1186/1747-5341-2-31http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2228286/Crossref Scopus (8) Google Scholar, 8Kelly A.R. Autry M.K. Access, accommodation, and science: knowledge in an “open” world.First Monday. 2013; 18 (Accessed June 3, 2013)http://firstmonday.org/ojs/index.php/fm/article/view/4341/3684Crossref Google Scholar, 9Poynder R. Open access by numbers Blog post, Open and shut?, June 19, 2011.http://poynder.blogspot.com/2011/06/open-access-by-numbers.htmlGoogle Scholar broadening access while maintaining standards through scrupulous peer review. The practice has spread to major subscription publishers as well: Elsevier (Annals of Emergency Medicine's own publisher), Taylor & Francis, John Wiley, and others allow OA to individual articles in perpetuity for a fee, an arrangement known as hybrid OA. Some commentators describe OA as not only a reaction to soaring subscription prices but also nothing less than the future of scientific communication. Yet the movement's quality control mechanisms are under attack during a critical stage in the formation of its institutions and their broader acceptance as a credible alternative to subscription publishing.The Blacklist on the Bleeding EdgeJeffrey Beall, scholarly initiatives librarian and associate professor at the University of Colorado Denver's Auraria Library, became concerned about the proliferation of enterprises that he has termed “predatory publishers” while serving as the editor of the Journal of Library Metadata, when he noticed increasing amounts of spam e-mail inviting submissions and positions on editorial boards, often for journals of dubious relevance to his work. He began collecting names of author-fee-supported publishers engaging in these and other questionable practices, posting them on his blog Scholarly Open Access and advising researchers and others to avoid doing business with them. “The scientific article market is flooded with these low-quality journals, many of which haven't gone through an honest peer review,” Beall commented. “They're just polluting scholarly communication.”Beall's list has grown from a small personal project to a more structured instrument with a comments section, publicly posted inclusion criteria,10Beall J. Criteria for Determining Predatory Open-Access Publishers (2nd edition) Scholarly Open Access, 2012.http://scholarlyoa.com/2012/11/30/criteria-for-determining-predatory-open-access-publishers-2nd-edition/Google Scholar and a formal appeal process allowing publishers to rebut the contention that their journals have earned this badge of dishonor. (At least 2 high-volume journal publishers, Dove Medical Press and Hindawi, have been removed after appeal and reanalysis.) Hundreds of organizations and publications have made the list to date by neglecting basic editorial functions (not only peer review but also minimal copyediting), lacking or faking editorial and review board expertise (obscuring their officials' identities or using names without permission), demonstrating a lack of operational transparency, invoicing authors for steep and undisclosed fees after accepting articles and requiring transfer of copyright, letting authors review their own articles under fraudulent online identities, confusing researchers by promoting conferences with names resembling those of respected scientific society events (separated from the recognized conference's name by a single hyphen in one case), plagiarizing material, publishing pseudoscience, and other offenses.“The predatory publishers chiefly operate in the areas where there's grant money,” Beall reported, “so it's the life sciences, for the most part. I can't think of any emergency medicine journals among these publishers … . [I]t's a very specialized topic, and most predatory publishers try to have broad scopes.”The titles of these enterprises are often general, bordering on meaningless, he said, and redundancy is a minimal concern: the list includes 2 firms (one Nigerian, one Ghanaian) using the name International Research Journals and one going by International Research Journal (based in India). Another listed firm, CSCanada (aka Canadian Research and Development Center of Sciences and Cultures), publishes the baffling title Higher Education of Nature Science, instructing prospective authors that “[a]rticles will include advanced knowledge, significant discoveries and the development of nature science at the universities and colleges, designed to lead to critical insights into the areas being addressed.”11Higher Education of Nature ScienceCSCanada [Higher Education of Nature Science Web site].http://www.cscanada.net/index.php/hensGoogle ScholarBeall's mentor William Cohen, founder of the Haworth Press (the Journal of Library Metadata's publisher, now part of the Taylor & Francis Group), described these firms as boiler room–style operations with little incentive to build serious reputations, unlike established publishers in either the subscription or OA realms. Though both Beall and Cohen compare the predatory publishers to vanity presses, the analogy goes only so far. A literary vanity press is a harmless indulgence; the ones on Beall's list introduce unvetted claims into the scientific record on a scale that does nontrivial damage to its credibility. Of the publishing houses targeted by Beall, Cohen said, “each one launches between 1, 2, or 3 to a dozen, several hundred journals each. They just list whole spreadsheets of them all at once. It took me 30 years to launch 200 journals at Haworth; some of these launch 200 in a day.” Arguments that the volume of research being conducted in biomedicine and other fields has expanded enough to require so many new outlets, he added, are “bollocks …. It's well known in publishing that only a small percentage of the journal articles are ever cited or used.”Some commentators see the effect of the junk journals amplified beyond their actual scale. Biophysicist Cameron Neylon, PhD, advocacy director for PLoS, calls them “relatively small operations” but finds that the scammers' conduct means that for OA publishing in general, “everyone gets tarred a bit with the same brush … . There's an awful lot of dodgy behavior happening in subscription space, and people don't seem to pay any attention to that, but that's the nature of being the new people on the block.”The junk publishers, in the PLoS perspective, are “the equivalent of a Nigerian e-mail scam. It's at that level of sophistication in most cases, and it's at that level of obviousness. It's not so much that it's a real problem; it's that it is damaging for the industry as a whole.”Slapped for Speaking UpBeall recognizes that publicizing the shadow world of predatory publishing makes him a flak magnet. “I think everything that I've done is protected by free speech rights,” Beall said. “I review books, and I write negative and positive book reviews. I see my work with predatory publishers as basically along the same lines. It's just like book reviewing, saying what's good and bad about something. It's just an extension of critical thinking.”Regardless of the salutary effects of this whistle-blowing, however—or the freedom of speech implications—enterprises included on Beall's list have fired warning shots across his bow.A pair of attorneys speaking for the Hyderabad-based OMICS Group recently sent Beall a demand for $1 billion in damages, plus an additional $10,000 fee merely for the alleged cost of preparing this demand itself, alleging that including the firm on his list (where it has been a focus of complaints well beyond Beall's own analyses) was libelous. The threatening letter combines a barrage of intemperate, often ad hominem accusations with nonspecific evaluative defenses of the OMICS Group's quality of work. Another aggrieved party, the Canadian Center of Science and Education, has threatened litigation over the listing of its 4 journals; the owner is a nonscientist Toronto proprietor with a history of involvement in plagiarism.12Brean J. Plagiarized reports from Iranian researchers printed by Canadian publisher.National Post (Toronto). December 3, 2004; (Accessed June 4, 2013)http://news.nationalpost.com/2012/03/04/iranian-researchers-plagiarized-material/Google Scholar These actions fall under the description of “strategic lawsuits against public participation” (SLAPPs), a common tactic for organizations aiming to silence a critic through fear of high legal expenses.“I've received these before,” Beall responded. “I don't think that … using threatening letters as a business model is very effective in the long run.” Nevertheless, at this writing, his university's legal office is reviewing the threat.He emphatically rejects the contention by OMICS lawyers that his gatekeeping reflects bias against researchers from developing countries, noting that playing the ethnic prejudice card shortchanges the many researchers from those nations whose work appears in legitimate and prestigious journals.“I don't think the western publishers are ignoring scholarship that comes from nonwestern countries,” he commented. “I think they're doing a very good job of attracting the best scholarship, and I think that the scholars in developing countries are just as smart as the scholars in any other country, and they can submit articles that are just as innovative as anybody else can.”Another glaring inaccuracy in the OMICS letter, he noted, is the charge that he wants “to strangle the culture of Open Access publication and would want to propagate the paid access publications.”13Letter from Ashok Ram Kumar and Vijay Kumar, Attorneys, IPmarkets, Hyderabad, India, to Jeffrey Beall, May 14, 2013.Google Scholar Beall has recurrently stated his views that legitimate OA is “a social movement that has changed academic publishing for the better, lowered costs and expanded worldwide access to the latest research”14Beall J. Predatory publishers are corrupting open access.Nature. 2012; 489 (Accessed June 4, 2013): 179https://doi.org/10.1038/489179ahttp://www.nature.com/news/predatory-publishers-are-corrupting-open-access-1.11385Crossref PubMed Scopus (487) Google Scholar and that he is “not ‘anti’ anything [but] in favor of the best model for scholarly communication, whatever it turns out to be.”15Poynder R. The OA interviews: Jeffrey Beall, University of Colorado Denver Open and shut?, July 11, 2012.http://poynder.blogspot.com/2012/07/oa-interviews-jeffrey-beall-university.htmlGoogle ScholarStevan Harnad, PhD, professor of cognitive science at the University of Southampton and at Université du Québec à Montréal, founding editor of the journal Behavioral and Brain Sciences (an innovator in using open peer commentary), and a recognized OA pioneer, sees such threats as nuisances meant to intimidate through “FUD: fear, uncertainty, and doubt.” The letters to Beall remind Dr. Harnad of threats lodged against British journalist Richard Poynder in 2008 over his investigations into Scientific Journals International, a publisher linked with spamming and an unusual fee structure in which article processing charges varied with the number of authors on an article's byline. Dr. Harnad aided in a legal defense fund for Poynder at the time, though it ultimately wasn't needed.“I don't think anything that Jeffrey has said about them is litigable,” Dr. Harnad continued. “He says, ‘Because of A, B, and C I draw conclusion D,’ and he tells you what A, B, and C are, and those are objective, and conclusion D is his summary of what this means … . Anybody that knows peer review knows that these junk shops are not doing peer review.”Metrics, Watchdogs, and Indices Versus Gresham's LawConsidering the apparent aggressiveness of some of Beall's targets, Cohen described him as “a very brave person to do this … . The only safeguard he's got is, no one's going to go after him; there's nothing to get. He's not rich.”Should the threats proceed to actual litigation in a US, Indian, or Canadian jurisdiction and somehow either succeed—unlikely on a prima facie basis, considering the quality of the claims and, in OMICS' case, the self-parodying amount sought—or, conversely, fail but generate daunting legal costs, a chill could fall over communications about scholarship and pseudoscholarship.Cohen concurred with Dr. Harnad that this, not a billion-dollar windfall, is the real aim of the threats. Proving in court that a publisher's editorial procedures meet or do not meet scientific standards, he said, would be “murky … . There's no universal recognition of acceptable professional peer review.” Because the quality and prestige of journals are vital to academic career assessments, he added, a silencing of Beall's list might also hinder the ability of academic administrators to evaluate researchers' publication records.Here, Beall's efforts complement existing mechanisms and compensate for some of their limitations. Peter Suber, PhD, JD, director of the Harvard Open Access Project, faculty fellow at Harvard's Berkman Center for Internet and Society, senior researcher at the Scholarly Publishing and Academic Resources Coalition, open access project director at Public Knowledge, and research professor of philosophy at Earlham College, noted that quantitative metrics of journal quality such as citation indices, H-indices, or journal impact factors “allow non-experts to judge experts—which is a good thing.” Yet these metrics can be inaccurate and can be gamed. Poor articles, such as the irreproducible, widely disparaged cold-fusion claims of Stanley Pons and Martin Fleischmann, often have high citation indices because they are repeatedly cited as punching bags. Colleagues engage in log-rolling reciprocal citation, a practice Dr. Suber finds “almost invisible to the world.” Some journals even insist that authors cite other authors from the same journal; “it's unethical, but it happens, and it happens precisely because the impact factor is the coin of the realm.”“Quality is either impossible to measure,” Dr. Suber summarized, “or we haven't figured out how to measure it so far, so we fall back on quantitative parameters that we can measure. It's the contact lens problem: if you lose your contact lens, you look for it under the streetlamp, not because you dropped it there, but because that's where the light is … . I understand why they've turned to the impact factor, but we give it too much weight; we treat it as a quality metric, which it never was.” (He contrasts impact factors with citation systems in law, which count and distinguish citations that approve, disapprove, or overrule the case cited; something similar in scientific scholarship would make partial headway against distorted metrics, though to scale up appropriately, it would require a workforce as diligent as the armies of law students who read and annotate cases.)Lax editorial standards carry more broadly consequential risks, Cohen observed, because scientists are not the only consumers of scientific literature. “The question ‘Who uses and cites journal articles?’ affects public policy,” he said. “Lawyers do, judges do, doctors do, government officials do … congressional committees … a news station. That's what scares me.”In an online publishing environment in which a version of Gresham's Law—the economic principle that a lower-quality currency tends to crowd out a higher-quality currency because the latter is hoarded rather than circulated—appears to apply, Beall's investigations and conclusions provide handy filtering tools.The binary simplicity of the blacklist, however, calls for caveats. “I would distinguish between low-quality journals and dishonest journals,” cautioned Dr. Suber. “They're both problems, but they're separate problems. And the same distinction exists on the subscription side: there are low-quality subscription journals and there are dishonest subscription journals. Some of them are in both categories at the same time. I think Beall's list doesn't distinguish low quality from dishonesty.”The challenge, according to Dr. Suber, is to sort out genuine scam artists from “good-faith startups who aren't trying to deceive or exploit or rip off anybody, but who aren't publishers and don't know the professional standards of publishing. I think for them, the solution is not condemnation or exile, but assistance, especially from publishers who do know the standards of the profession.”A buddy system pairing newcomers with experienced publishers, he suggests, might help startup editors and publishers improve. He expressed a concern that “if the list is not scrupulously assembled, then some innocent publishers will be stigmatized as predatory, and that's a more serious problem than some dishonest ones escaping for the time being.” Beall's list, he speculated, might also serve its purposes well by expanding beyond Beall's own efforts. “Right now, it's Jeffrey Beall's judgment,” Dr. Suber commented, “and his judgment is pretty good, but maybe communal judgment would be better.”Another guide for researchers or readers seeking authentication of unfamiliar journals takes the form of “whitelists”: compendia of reliable OA journals. The traditional criterion for journals, indexing by the former Institute for Scientific Information (ISI, now the Thomson-Reuters Web of Science), is still valuable, but “probably overexacting,” Dr. Harnad said, “because there are probably third world journals that are not indexed by ISI but are still OK.” The expanding OA sector has outpaced this index's growth; the OA community has also shot itself in the foot, Dr. Harnad believes, by “trying to assert itself prematurely.” Claiming “prejudice in favor of the journals that are indexed by ISI and their famous impact factor,” he said, was “one of the stupid things that people who are trying to promote open access said which they shouldn't have said.” He advises beginning researchers to take ISI/Web of Science indexing seriously, avoid the “rah-rah OA” temptation to dismiss that index as behind the times, and aim for “the highest-quality journal whose standards you can meet.”Some junk journals, Cohen added, have made misleading claims about being included in another list, the Directory of Open Access Journals, launched at Lund University in Sweden. This project aims to provide a comprehensive global list of OA journals practicing peer review or editorial quality control, and it is gearing up to perform that function, but according to both Cohen and Dr. Suber, it does not yet have the resources to investigate journals' practices, track them over time, and provide the equivalent of a Good Housekeeping seal. (Some organizations, Cohen added, simply lie about being included there or in familiar services such as MEDLINE, PubMed, or PsycINFO; cross-checking to verify claims is both advisable and time-consuming.)A more rigorous imprimatur, Dr. Suber said, and one not yet on the radar screen for many researchers and physicians, is membership in the Open Access Scholarly Publishers Association (OASPA). This is a professional society based in The Hague, with a code of conduct, a preliminary investigation for prospective members, expulsion procedures if a member fails to comply, and a staff. Dr. Suber, although recommending direct examination of a journal's articles and masthead (perhaps augmented by consultations with informed colleagues) as the best means of evaluation, added that scholars can regard OASPA members as trustworthy.“Now, the problem is the other side,” he cautioned: “if [a publisher] doesn't belong to OASPA, it may not have gotten around to it yet. There's a long tail of open access publishers, hundreds, and most of them don't belong to OASPA yet, but I'd like to create an incentive for all open access publishers to join OASPA and therefore to be bound by this code of conduct … . It's self-regulation by open access publishers. Everything that open access's opponents can say in criticism of predatory journals, basically OASPA has already said.”Cohen suggested 2 other reforms that could help researchers and readers check under the hood: a slight change in standard bibliographic formatting and the creation of “a union list of primary indexing and abstracting services.” If citations to journal articles listed an item's publisher, as well as the journal title, adapting the convention already used for books, junk publishers would be more visible. A regularly updated master list merging the major indexing and abstracting services in different fields would allow authors to “plug in the name of a journal and find out where it is actually indexed, not where a publisher claims it is indexed.” (Ulrich's Periodicals Directory, Cohen reported, attempts to accomplish this, but in his early career, filling out Ulrich's forms, he found that they simply accepted publishers' entries and lacked staff capacity to check them. An effective union list, he speculated, would require “a mammoth effort without producing any income.”)At the opposite end of a scale of rigor and responsibility lies a type of intervention that combines the purposeful subterfuge of Alan Sokal's Social Text hoax16Sokal A.D. Transgressing the boundaries: towards a transformative hermeneutics of quantum gravity.Social Text. 1996; 46/47 (Accessed June 4, 2013): 217-252http://www.physics.nyu.edu/faculty/sokal/transgress_v2/transgress_v2_singlefile.htmlCrossref Google Scholar, 17Sokal A.D. A physicist experiments with cultural studies.Lingua Franca. May/June 1996; (Accessed June 4, 2013): 62-64http://www.physics.nyu.edu/faculty/sokal/lingua_franca_v4/lingua_franca_v4.htmlGoogle Scholar with the snarky flair of an MIT student prank. Independent researcher Phil Davis, using text-generating software (SCIgen18Stribling J. Krohn M. Aguayo D. SCIgen—an automatic CS paper generator.http://pdos.csail.mit.edu/scigen/Google Scholar) that produces grammatically correct but “content-free” articles, fitted out with figures, tables, and references—in other words, gibberish plausibly formatted to resemble research reporting—submitted such an article to a fee-supported journal in computer science, the Open Information Science Journal, which claimed to practice peer review. Despite its nonsensical content and a reference to its imaginary coauthors' institutional affiliation with a Center for Research in Applied Phrenology (CRAP), the article was accepted (with an invoice for an $800 fee).19Davis P. Open access publisher accepts nonsense manuscript for dollars.Scholarly Kitchen. June 10, 2009; (Accessed May 30, 2013)http://scholarlykitchen.sspnet.org/2009/06/10/nonsense-for-dollars/Google Scholar Similar applications of SCIgen, producing work purportedly authored by one Herbert Schlangemann, have yielded not one20Let's generate a paper Blog post, The Official Herbert Schlangemann Blog, June 1, 2008.http://diehimmelistschoen.blogspot.com/2008/06/lets-generate-paper.htmlGoogle Scholar but several acceptances of articles for conferences, including a reviewed Institute of Electrical and Electronics Engineers event, plus the selection of the nonexistent and crudely christened Dr. Schlangemann as the session chair.21Newitz A. Computer-generated paper accepted for prestigious technical conference Blog post, io9, December 24, 2008.http://io9.com/5117892/computer+generated-paper-accepted-for-prestigious-technical-conferenceGoogle ScholarBeall and Cohen do not recommend SCIgen-enhanced exposure; it requires outright lying in the standard warrant that an article represents real and original work. Still, they recognize the point behind the pranks: even reputable professional societies, like Homer as Horace described him, sometimes nod. If review standards can occasionally lapse to the point of letting a Schlangemann past the gates, then, a fortiori, organizations that pay lip service or less to peer review require vigilant skepticism.(Humility and full disclosure require a recognition that established publishers, too, make egregious mistakes: Elsevier, Beall noted, let its standards lapse in publishing several pharmaceutical company–sponsored journals in the early 2000s without disclosing the conflict of interest.22Grant B. Elsevier published 6 fake journals.Scientist. May 7, 2009; (Accessed May 29, 2013)http://www.the-scientist.com/?articles.view/articleNo/27383/title/Elsevier-published-6-fake-journals/Google Scholar “They got caught, and they apologized for it, and it's done,” he said, though not without reputational damage.)PLoS PlusPLoS has demonstrated that OA journals can attain high prestige and strong impact metrics without requiring exclusive rights. Though commercial publishers have argued that exclusivity is essential to maintain high editorial standards, PLoS has done so—and made a surplus, Dr. Suber noted; “it's nonprofit; otherwise, we'd say it was making a profit”—while publishing all its content OA. Major US policy initiatives may accelerate this movement: the Presidential Office of Science and Technology Policy directive of February 22, 2013,23Holdren J.P. Memorandum for the heads of executive departments and agencies Office of Science and Technology Policy, February 22, 2013.http://www.whitehouse.gov/sites/default/files/microsites/ostp/ostp_public_access