Scientific publishing has undergone dramatic and rapid change in the past decade. With the launch in 2003 of the PLoS juggernaut, founded on the idea that editors and referees should evaluate a manuscript only for its soundness and that publication charges should be paid directly by the author (replacing the subscription model), the genie was out of the bottle. The intoxicating notion of grassroots science, viewable to all via open-access (OA) publishing, was impossible to resist.
That publication model has now been duplicated by virtually every major scientific publishing house in the world, including those of professional societies. The two biggest scientific megajournals, PLoS One and Scientific Reports, churn out more than 10 000 publications per quarter in all the STEM fields. Most observers would agree that academic publishing has changed forever.
The rise of OA and the megajournals has turned out to be a lucrative model for publishing houses.1,2 But is it good for the scientific community as a whole? Opinions on this differ from field to field, with the more translational fields, like biology and medicine, taking a more enthusiastic stance and more fundamental fields, like mathematics and physics, a more skeptical one. (See the commentary by Jason Wright in Physics Today, February 2020, page 10, and reference 3.)
There is also a noticeable generational difference of opinion. Some younger scientists view the trend toward OA scientific journals more favorably than their older colleagues do. They may be more likely to embrace the implicit goal of this new publication model, which is to get more people around the world to view and cite their papers so that their h-index3 and other bibliometric indices can grow. After all, who would be content with a measly h-index of 60, like that of the venerable Richard Feynman, when academic “influencers” in certain fields can now drive their indices well above 100, or even 200? Such numbers would confirm their clout and impact, boosting their h-index further in a never-ending feedback cycle.
But all is not perfect with the new OA model. First, whereas virtually everyone endorses OA’s laudable goal of making articles free and available worldwide (not just to elite academics with institutional subscriptions), the model relies on article publication charges paid directly by the author. The charges can be significant, say $4000 or more per article. A direct payment link between author and publisher replaces the previous subscription model between institution and publisher. In the new model, unlike the old, the more articles published, the more the publishing house profits. Quantity replaces quality.
That switch has opened the door to huge growth in the number of low-quality OA journals, supported by author-paid publication fees comparable to, or sometimes exceeding, those charged by the top OA journals. Those so-called predatory journals are proliferating.4
To get an idea of the scope of the problem, Cenyu Shen and Bo-Christer Björk of the Hanken School of Economics in Helsinki, Finland, documented an increase in publication volume from journals they categorized as predatory. In particular, they found an increase in papers from 53 000 in 2010 to 420 000 in 2014, with 75% of the corresponding authors from India, other Asian countries, and Africa.5 How, we wonder, can authors from around the world pay the article-processing charges?
Shen and Björk’s findings also support our suspicion that charging fees motivates journals to accept more articles. A further incentive comes from governmental funding agencies (supported by taxpayer dollars), which now require their funded scientists to publish in OA journals (even if their manuscripts are first posted on preprint servers free of charge and available to all readers). The requirement seriously undercuts prestigious and established journals that do not charge publication fees and, ipso facto, are not OA.
That threat brings us to our second point. The nonjudgmental, accommodating editorial policies of most OA journals and the proliferation of low-quality journals have worried many scientists who have watched the publication trends for 20 or more years with some alarm.
To make matters worse, many publishing houses have now adopted the model of having one “flagship” journal, which attracts the bulk of the manuscripts and publishes the best of them, as well as several lesser journals ready and willing to publish many of the rejected articles from the flagship in a trickle-down process that seems to leave no article (or publication charge) behind. The desire of publishers to publish high-quality content even if it might not be appropriate for their flagship journal is understandable. But the incentive to hang on to potential publication fees exists as well. If an author is willing to pay the fee, surely the free-market system incentivizes the publisher to accept that fee. That direct link between buyer and seller interferes with the scientific evaluation process.
Our third point concerns the impact on reviewing. The huge increase in journal publication numbers ushered in by the OA model makes it very difficult to obtain meaningful, credible, and expert referee reports on articles submitted to all journals.6 The reports are crucial not only to flag articles that should not see the light of day, but also to suggest improvements to articles that present significant results yet are unpolished or poorly articulated. The vast majority of articles published over the years under the traditional model have benefited greatly from this important iteration step, guided by experts. If the system implodes, quality will decline further, and the credibility of the scientific enterprise as a whole will inevitably suffer.
Authors are being placed in the unenviable position of being simultaneously caught on the horns of a dilemma while being forced to accept a Faustian bargain. The dilemma is to choose between funding a new incoming graduate student eager to be trained or paying the publication charges on papers produced by advanced graduate students anxious for publications. The bargain is that by attempting to push up their h-index with exposure to a wider audience, they are in danger of lowering their publication standards and hurting their scientific reputation.
The new scientific publication model that has taken hold over the past decade is evolving, and we, as scientists, have both the opportunity and the obligation to influence it. Gone is the era (some would argue for the better) when well-meaning editors dominated a field and shaped it according to their taste and style. Crowdsourcing and impact factors are replacing the careful curation and development of ideas. Many of us entered our profession driven by intellectual curiosity, with the belief that doing good science would be our ultimate reward. This, unfortunately, seems to have carried with it the danger of leaving us vulnerable to exploitation.
We are in the middle of a publication pandemic that needs to be managed, and all of us must play a role in trying to stem that tide. We were all influenced by inspiring and creative papers when we were graduate students. Take the time, once again, to glance at those papers and other landmark papers written by the most influential scientists in your field. Ask yourself why these papers were memorable, and almost surely it is due to their originality and readability. Those papers continue to set an upper bar on what is achievable and should be what we strive for. When you listen to talks at conferences, most likely the ones that stand out are ones that are original and direct, not ones that string together ideas from long lists of publications. Publishing in journals that are most suited for getting your specific ideas across, run by editors who are practicing scientists with domain-specific expertise, is something to keep in mind when selecting where to submit your articles. We can all create a culture in which publication quality matters more than publication quantity, no matter the stage in our careers.
To the extent that researchers are publishing promiscuously, it must be recognized that they are motivated to do so by the rewards that numbers seem to bring. To the extent that practicing scientists can change this motivation, we must try to do this consistently by reasserting the notion that creativity matters more than citation statistics. We must work with the publications industry to make our voices heard in order to steer the system in a direction that ensures our mutually beneficial existence and not our mutually assured destruction.
If we do not assert ourselves now, we will be viewed as collaborators in a downward intellectual spiral that is accelerating. After all, to paraphrase the words of Eugene Wigner,7 the miracle of our scientific enterprise, which exists symbiotically with the scientific publications industry, is a wonderful gift we neither understand nor deserve. We should be grateful for it and hope that it will remain viable well into the future.
- D. Butler, Nature 495, 433 (2013).
- R. Van Noorden, Nature 495, 426 (2013).
- J. E. Hirsch, Proc. Natl. Acad. Sci. USA 102, 16569 (2005).
- A. Grudniewicz et al., Nature 576, 210 (2019).
- C. Shen, B.-C. Björk, BMC Med. 13, 230 (2015).
- J. Bohannon, Science 342, 60 (2013).
- E. P. Wigner, Commun. Pure Appl. Math. 13, 1 (1960).
Paul Newton is a professor of aerospace and mechanical engineering and mathematics at the University of Southern California in Los Angeles and currently serves as editor-in-chief of Journal of Nonlinear Science (Springer Nature). Katepalli Sreenivasan is a professor of physics and mathematics and the Eugene Kleiner professor for innovation in mechanical engineering at New York University in New York City and is former editor-in-chief of Journal of Nonlinear Science.