Bookmark and Share

Open access open for business

Massimo Nespolo
[OAmoney]

Publication of scientific research results requires a validation process, without which articles would not be more trustworthy than blog posts. Although certainly imperfect, like any human activity, the best validation model we have been able to develop so far is the peer-review process, in which a panel of (at least two) anonymous reviewers (“referees”), solicited by an editor, critically analyse a manuscript submitted for publication. The reviewers' task is to find any possible flaw, mistake or overlooked material in the manuscript that may have gone unnoticed by the authors. It is a well-known fact that authors are often blind to their own mistakes; the good practice of putting a manuscript asleep for a couple of weeks before submission and reading it once again after this rest period helps in seeing that which has escaped at the editing stage. Unfortunately, in the world of “publish or perish”, delaying a submission has become a luxurious practice seldom adopted, especially when working in a highly competitive field.

Publication of reviewed manuscripts was in the hands of scientific not-for-profit societies until commercial publishers understood that it represented a lucrative market and joined the business; this change brought in a revolution in terms of technical handling and customer-oriented approach, but handling of manuscripts was still the responsibility of renowned experts in the field. Access to published manuscripts required one to pay a physical visit to a library. Later on, on-demand copy services were introduced, which eased somewhat the task of getting one's hands on the articles. With the exponential increase in published articles, libraries started to have space problems and had to move older issues of their journals to storage facilities, or to microfilm them. At the same time, the cost of subscription to the ever-increasing number of journals skyrocketed, and the profits of commercial publishers kept on growing. The sustainability of the model began to be questioned. An ever-increasing proportion of the results of research, funded in most cases by tax-payers' money, finished in the hands of commercial publishers and researchers had to use further tax-payers' money to access these results. Choosing not-for-profit scientific societies like the IUCr would have limited these costs; unfortunately, a vicious circle arose (or rather, was put in place): evaluation of research results and decision of what research to fund moved into the hands of non-specialist politicians who require “objective” criteria in some numerical, easy-to-read form: the journal impact factor is the most visible and best known indicator. Commercial publishers, especially large companies, have enough market power to promote their journals and obtain more visible (“quotable”) articles; once a high-index profile has been established, it is easier to keep on attracting submissions on “hot topics” that preserve the high-index profile. Competition is not easy for not-for-profit scientific societies, although some of them, like the IUCr, RSC and ACS – just to mention a few – do a remarkable job.

With the move to electronic-only publication, the printing costs have been cut down, together with the publication delay. At the same time, the proportion of the job done by the authors themselves has increased. Unlike the IUCr, some commercial publishers do not assure a high-level technical editing process but require files almost ready for electronic archiving. The high costs of the subscription to electronic-only journals that are commercially published are therefore even less justifiable.

This approach has been challenged by the Open Access initiative (formulated in 2002 in the Budapest Open Access Initiative), which in principle tends to move the costs of scientific publishing from the reader (subscription) to the author. Well before the term “open access” came into use, computer scientists started the practice of self-archiving in anonymous ftp archives; later, physicists followed the same path by using arXiv, a repository of electronic preprints that are accessible to the general public after moderation. This approach has at least two drawbacks: for the community, because “moderation” is by no means a guarantee of quality comparable to peer-review (an extreme example can be seen at the address https://arxiv.org/pdf/0906.1358); and for the authors, because a “moderated” (non-peer-reviewed) manuscript is not recognized as a research result. These drawbacks are overcome by the possibility of publishing open-access articles in existing journals, where both classical (subscription or pay-per-article) access and open access coexist, although this practice requires the payment of open-access fees. IUCr journals have been among the pioneers along this route, the first open-access article being published in 2004, with approximately 30,000 open-access articles published up to the end of 2018. In addition to seven hybrid journals, the IUCr today publishes three entirely open-access journals: Acta Cryst. E, the high-profile journal IUCrJ, as well as IUCrData.

The current situation is particularly complex and financially challenging because of the co-existence of the two models (subscription and open access). Libraries still have to pay for highly expensive subscription packages, which represent a very important source of revenue especially for large commercial publishers. At the same time, an increasing number of journals are moving towards an open-access-only submission. Considering the increased number of views that an open-access article receives with respect to a traditional one, and the fact that some high-impact journals have adopted this model, the pressure on researchers to publish in open-access journals is increasing, sometimes to the point of jeopardizing the (constantly shrinking) recurrent research budget (for those institutions that still provide some budget besides the grants for which one has to apply). One may therefore imagine that the solution would be to move to a fully open-access model: the budget currently devoted to library subscriptions would be transferred to paying the open-access fees, possibly saving on the global expenses. Such a simple position does not take into account a serious threat that has arisen with the open-access initiative: predatory publishers.

Wikipedia has a precise definition for this phenomenon. “Predatory open-access publishing is an exploitative open-access academic publishing business model that involves charging publication fees to authors without providing the editorial and publishing services associated with legitimate journals”. The “editorial and publishing services” mentioned in this definition concern not only the quality of technical editing – which is limited or absent in predatory journals – but especially the peer-review process. The source of revenue being the open-access fees paid by the authors, each manuscript rejected represents a loss of revenue. Moreover, each manuscript that is delayed because of extensive revisions requested by the referees represents a delay in the payment, aggravated by the risk that the manuscript be withdrawn if the authors realize that they cannot comply with the requested revisions. The temptation to loosen the review process is constantly in the grip of less scrupulous publishers, which, brought to its extreme consequences, leads to by-passing it completely. The lucrative market of academic publishing has therefore been polluted by predatory publishers that accept whatever manuscript is submitted provided that the open-access fees are paid. Such a perverse phenomenon has reached extreme proportions today with the ease of starting an online journal through ready-to-use software packages. One may spend a few hours in browsing the websites of open-access journals and look at the editorial board, often composed of totally unknown people who have no serious publication history in the field, or even at all.

Jeffrey Beall, a librarian of the University of Colorado, set forth criteria for categorizing publications as predatory and maintained a list of predatory journals. He was forced to take his list offline in January 2017, owing to serious legal threats by publishers appearing in his list. The list, however, has been taken up again and is now maintained “by an independent group which wishes to remain anonymous in order to avoid the harassment suffered by the creator/maintainer”: it is available at the address https://predatoryjournals.com/publishers/. Other initiatives to keep track of predatory publishers exist as well, like the Flaky Academic Journals website. Unfortunately, not only are these lists not (yet?) officially considered as a reference by the academic authorities, but the most dangerous predatory publishers succeed, with a clever strategy, in remaining borderline outside the list, as we are going to see: I will call them “insidious publishers” hereafter, for the sake of brevity. I will not try to mention any of these explicitly to avoid enduring what Jeffrey Beall had to suffer, but I am confident that most readers will easily recognise the targets. Before going more into depth, however, it is interesting to point out that several researchers have attempted, and sometimes have succeeded, in trapping predatory publishers by submitting completely nonsensical or even hoax manuscripts, that were accepted without any review process. A famous example is that of the Star Wars Hoax Paper, where mitochondria were called “midichlorians”: extensive quotations from George Lucas movies were included in the manuscript, and a large number of fake references were provided as “literature”, attributed to non-existing authors whose names are simply characters from the galactic saga. Hard enough to believe, the hoax manuscript was published in four journals. It has since been taken off-line but I could download a copy before it disappeared: for those who are curious enough to see how huge the hoax was, the file, with the Star Wars references highlighted, is now available at the address https://frama.link/dq1A1WGQ.

An opposite approach with respect to the predatory publisher list has been adopted by the DOAJ (Directory of Open Access Journals), which defines itself as “a community-curated online directory that indexes and provides access to high-quality, open access, peer-reviewed journals”. However, as clearly stated in the top page of the directory, “All funding is via donations, 40% of which comes from sponsors and 60% from members and publisher members”. Among the sponsors one can see institutions and publishers that are above any possible suspicion; but there are also a number of insidious publishers. As Henry Louis Menken wrote, “It is a sin to believe evil of others, but it is seldom a mistake”. Initiatives like DOAJ should be encouraged, but one needs to ensure that they remain completely independent in their assessments of journals.

Let us finally analyse how the insidious publishers succeed in sneaking into the market and what are the potential long-term consequences of their practice. The strategy of remaining outside the blacklists exploits a fundamental defect of the scientific mind: most scientists are fundamentally too trusting, inclined to think that their environment, colleagues and co-workers are basically honest and motivated by intellectual curiosity and rigour, which unfortunately is not always the case. The attitude “innocent until proven guilty” that we more or less unconsciously adopt as a forma mentis is being exploited by dishonest commercial practices, against which the opposite attitude “guilty until proven innocent” should be mandatory.

An editorial board composed of upright scientists would mean a high rate of rejection of low-quality manuscripts, and therefore a limitation on the source of open-access fees. An editorial board composed of unscrupulous characters playing a fictional role would mean that every manuscript submitted goes online and provides a source of revenue to the publisher, but at the same time would also be a short-cut to the blacklist of predatory publishers. Insidious publishers smartly exploit a mix of the two to maximize the economic benefits while remaining borderline outside the blacklists. Concretely, the strategy adopted can be described as follows. The editorial board is composed of three types of editors:

  1. confirmed scientists, who seldom if ever receive manuscripts: their role is to bestow a certain degree of prestige to the journal; they are victims of the “innocent until proven guilty” syndrome:
  2. young scientists, who see their role as an opportunity to boost their CV; these are fundamentally honest people who, however, lack the necessary experience for a critical handling of manuscripts and are unable to resist the pressure by the publisher to let dubious manuscripts go through the review process, even when the reviewers’ reports are less than positive;
  3. non-entities on the publisher’s payrolls, who just play a role.

This is, however, not sufficient to grab a significant part of market: some degree of recognition is necessary, in particular to obtain a more than negligible impact factor. To reach this target a simple but effective strategy is put in place: special issues in the hands of guest editors. Some of these special issues are entrusted to well-known scientists, who are unaware of the flaws of the journal. For other special issues, young and less-known scientists are recruited, exploiting the same psychobiological mechanism used for the editorial board: flattery and the chance of CV-boosting. As for every special issue, editors are responsible for finding contributing authors. Invitations coming from well-known colleagues are a gauge of seriousness; waiving of open-access fees is even offered in some cases: the lack of revenue from those manuscripts is a sacrifice to build a reputation whose return on investment eventually pays off. When the invitation comes from unknown names with minimal publication records in the field, suspicion should be mandatory; the “innocent until proven guilty” flaw acts more than one would expect, but not enough to allow a complete success of the strategy. As a result, several special issues of these journals are closed with a ridiculously small number of contributions: sometimes not even a single manuscript appears after the deadline. If you keep an eye on the list of special issues, you will realize that those which failed to attract any contribution at all suddenly disappear, as if they never existed.

In some cases, experienced guest editors, who had seriously done their job, discover that manuscripts they had rejected have actually been published. Some authors see their submitted manuscript accepted in a couple of weeks, without even getting any review report. Reviewers for other serious journals discover the manuscript they had rejected appears online for these insidious journals a few weeks later, without any change with respect to the version that had been rejected (personal experience). It goes without saying that in a journal where practically everything is accepted, one finds also good articles by serious authors who are unaware of the predatory behaviour of the publisher. These authors are, however, sometimes guilty of refusing to see the evil even when others warn them. On more than one occasion I have tried to raise colleagues’ awareness about the dangers of contributing to such a scheme, but only in a few cases have these efforts been successful. After all, if their manuscript is worth publishing, what is the problem in choosing a journal that also possibly publishes some rubbish? It is that by contributing to those journals, these authors bring them a patina of respectability that allows the publishers to remain borderline outside the blacklists, to gather more manuscripts that could and should have gone to established journals, and to inflate their market. In the long term, the consequences could be much more serious. Somebody might think that the following analysis is simply a dystopian illusion; they are unwitting victims of their naive syndrome.

Public research is funded by tax-payers' money. Funding agencies require fulfilment of various criteria for a research project to be funded; among these, potential benefits for society, but also a guarantee of quality and reproducibility of results. Predatory publishers are seriously jeopardizing the system, with their dishonest commercial practices. Once the trust in the academic publication is lost, there will be no reason to spend tax-payers' money to pay open-access fees. This may eventually result in a collapse of the open-access system itself; or may even seriously threaten academic publication as a whole. In fact, once serious and legitimate doubts are cast onto the credibility of the results appearing in publications that have somehow gained a veneer of respectability, the whole world of scientific publication can easily be discredited. At that point, predatory publishers will simply move away from a market that is no longer lucrative and invest in some other deceptive activities. There will then not be any reason for them to maintain the infrastructure where their online journals were published. The whole database of their publication may therefore disappear into the world of paperless publications; in this case everything is lost forever, not only the huge amount of rubbish that went through, but also the legitimate publications by scientists who have been too unaware to even consider such a possibility. This could not happen with not-for-profit publishers like the IUCr, or major established publishing companies. But even if such extreme, catastrophic consequences are not reached, once the model of scientific publication as we know it today is discredited, nobody can predict what the consequences are going to be.

Before such a dystopian future becomes a dark reality, we have a moral duty to alert especially the younger generations to how to avoid the traps of predatory and insidious publishers. Without any pretence of being exhaustive, we can provide at least some simple guidelines to be taken as good-sense “rules of the thumb”:

  1. Small commercial publishers, with a short history, should be carefully scrutinized before submission of manuscripts.
  2. Publishers located in developing countries may not have the financial and technical solidity for sustainable business: a number of predatory publishers have their HQ in those countries.
  3. Special attention should be paid to publishers whose HQ is in a developed country but whose representatives and technical staff are either in developing countries or in countries that have an established reputation of forgery.
  4. Journals with a long list of editors who have no reputation in the field are particularly suspect. It is easy to check this by going to the journal’s website and ask yourself: how many of these people are known to you?
  5. Journals publishing a large number of special issues with very few, if any, articles, are probably issued by “insidious publishers”, if not recognized predatory publishers.
  6. Being invited to join the editorial board of a journal, or act as guest editor, at the beginning of one’s career, before even getting some reputation in the field, does not necessarily mean that your first publication(s) have earned the spotlight: more likely, it is the result of a robot-scan of recent journal issues. Established scientists receive multiple invitations of this kind per day and are obliged to use spam filters to get rid of them.
  7. A published article normally shows the manuscript history (if it does not, the journal has to be considered as highly suspicious). A too-short interval between submission and acceptance is clearly a bad sign. With some rare exceptions, reviewing a manuscript takes time: the reviewers have to read the manuscript, possibly some of the literature cited therein, check the experimental data or the equations or the flow of reasoning, and all this while attending their normal duties (lectures, meetings, writing their own articles...). A quick report submission by one reviewer sometimes occurs; by both reviewers of the same manuscript it is rare; by both reviewers of many manuscripts it is unrealistic and should raise suspicions. An average of – say – a 10-day review process for a journal simply means that the reviews are either non-existing or superficial and untrustworthy.

Insidious publishers have also started sponsoring small meetings and schools as a way to buy some credibility. Is that philanthropy? The amount of money they spend on having their logo on the event website, and to have the right to have their leaflets exposed during the event, is often inferior to the open-access fees of a single manuscript. One submission from a participant would be enough to cover the expenses. But even if no immediate economic gain is obtained, the return on investment in terms of image is more than worth the small contribution. Organizers of these events should be aware that by accepting this type of sponsorship, they indirectly contribute to feeding the parasite that is seriously threatening the world of academic publication as we know it.

10 February 2019