Scientific publication – Is it for the benefit of the many or the few?

Since the end of the 17th century, new scientific knowledge has been disseminated mainly through scholarly journals, usually controlled by researchers grouped within learned societies, such as the Royal Society of London and its journal founded in 1665.

Scholarly journals are usually managed by editorial committees that define the editorial policy and control the process of independent evaluation and revision of articles submitted for publication. Although there are many national scholarly journals, their content generally has an international scope, which is reflected in the international composition of the editorial boards.

Especially after the Second World War, scientific journals grew exponentially in number and diversity and private companies saw it as a very lucrative market. Firms like Elsevier, Springer and Wiley have thus become giants of scholarly publishing that more or less monopolise the most internationally recognised journals. All the strictly scientific work is still done for free by researchers, but the profits generated by the sales of these journals to university libraries are privatised in the hands of a few conglomerates.

All of this is relatively well known. What is less so, however, is the fact that some researchers have also understood the advantages of proposing to these giant publishers the creation of new journals that they then ‘manage’ in order to benefit their own research and that of their friends.

The El Naschie case

A striking example of such a strategy is provided by the Egyptian engineer Mohamed El Naschie, who was editor of the very specialised theoretical physics journal Chaos, Solitons & Fractals. It was created by El Naschie in 1991 and was then published by the Pergamon group, a publisher bought by Elsevier in 1992.

While working on a quantitative study of a particular scientific controversy, Yves Gingras was struck by the fact that El Naschie had published, between 1991 and 2008, nearly 269 articles in this single journal, representing more than 85% of its total scientific production during the period.

In addition, according to data from the Web of Science database, his articles have hardly ever been cited outside of the journal itself. We did not, at the time, see fit to alert the scholarly world of that curiosity. However, researchers eventually discovered it for themselves, causing a scandal in 2008 in the journal Nature.

In that year alone, El Naschie had written 53 articles in Chaos, Solitons & Fractals, raising serious doubts about its evaluation process. Indeed, it should be recalled that no serious journal publishes as many articles by its editor or even by any single author in a single year.

In these many papers, El Naschie essentially exposed the results of a theory he had developed suggesting that the universe has an infinite number of dimensions. Experts called upon to comment on this theory have called it either inconsistent or “undisciplined numerology full of impressive buzzwords”.

Faced with an outcry by scientists, Elsevier announced that El Naschie would retire in 2009 – an elegant way to “thank him” – and that a new editor would be found to reorganise the journal on a new basis. El Naschie then sued Nature for defamation, but his claim was dismissed in 2012.

Other ways to exert editorial control

This little-known example sheds light on ways academics can take control of a journal to facilitate the publication of a group of researchers. El Naschie did so in a very visible, and even simplistic, manner as editor-in-chief. However, there can be more subtle ways to control a journal without officially appearing on its editorial board, such as by being part of a local network of scientists that dominate the board of editors and thus control the journal.

A recent Médiapart journalistic investigation into the publications of Didier Raoult has drawn our attention to the journal New Microbes and New Infections, some of whose characteristics appear to us to be similar to El Naschie’s journal.

We therefore carried out a bibliometric analysis of this biomedical journal, listed in the Scopus database produced by Elsevier (which also publishes the journal). Founded only in 2013 – a period that saw the multiplication of new journals by large groups, with the aim of increasing the publishers’ revenues and diversifying their scholarly portfolio – New Microbes and New Infections produced 743 articles (as of 10 June 2020).

What is striking for a journal claiming “to cover almost the entire scientific world” is the fact that the countries which publish the most in it are: France (N = 373), Saudi Arabia (N = 115), Iran (N = 48), Senegal (N = 46) and Italy (N = 44). There follows a tail of countries contributing with very few articles since the creation of the journal.

France therefore represents 50% of total articles, while it has produced only around 7% of global virology publications between 2013 and 2020, as compared to 41% for the United States. Contrary to what the content of New Microbes and New Infections suggests, France is far from dominating the international field of the study of microbes and viral infections.

Now let's look at these French articles. First, we observe that 337 contain at least one institutional address of researchers based in Marseilles, that is 90% of the French total. By focusing on the authors, we then find that 234 of them, or two-thirds, are co-signed by the researcher Didier Raoult, who has recently been making media headlines for his controversial publications on the use of hydroxychloroquine to treat COVID-19 patients.

We also observe the rapid rise of this author in the journal: from a single article published in 2013 to a peak of 77 articles in 2017 alone. For the first half of the current year (as of 10 June), he already has 12 papers in the journal. In addition, the journal’s assistant editor, Pierre-Edouard Fournier, has published 170 papers in the journal since its creation.

Since scientific publications are normally peer reviewed and the decision to publish or not is made by a supposedly independent scientific committee, let us now take a closer look at the composition of the journal’s editorial board.

The editor is based in Marseilles, and among the 15 members of the board, six others are also French and five are from Marseilles, the other being from Paris. The ‘international’ nature of the board is still ensured by the presence of nine other members, from the United States (four), Algeria (one), China (one), Switzerland (one), Australia (one) and Brazil (one).

Although all the papers are supposed to be evaluated by independent external specialists, chosen by the editors of the journal, the fact remains that the strong local component – that is to say almost half of the total – of its editorial board may help explain the dominance of very local publications in this so-called ‘international’ journal. We do not know of serious scientific journals that accept that almost half of their editorial board of scientists are concentrated in the same city.

Keeping a critical eye on publications

This rapid bibliometric study of the dynamics of publications in scientific journals shows that a scientific analysis of publications can go much further than the very problematic uses made of them to ‘evaluate’ researchers and calculate their ‘h index’, considered by many naïve scientists to ‘measure’ the ‘quality’ of their research.

A good use of bibliometrics sheds a unique light on the sociology of science. It thus makes it possible to suggest that journalists who cover research in health, and more broadly in the sciences, should not be content with repeating the usual expression “published in a scientific journal”, but should also scrutinise the nature of the journal that announces the results they consider worthy of appearing in the mass media.

They should check whether these journals are the work of independent learned societies – for example, the journal Science is the property of the American Association for the Advancement of Science – or of private groups listed on the stock exchange.

They should also ask themselves whether the aim of these groups in rapidly publishing certain articles is not simply to maximise the visibility of their journals and, thus, increase subscriptions from university libraries.

Researchers should also question the relevance of constantly creating new scientific journals when the existing ones are more than sufficient to publish the most robust, useful and interesting results.

Keeping high selection standards helps to rule out questionable studies that are botched or carried out in a hurry, for instance, because an author is seeking job security or a research grant. By being selective, the best journals can also contribute to slowing down the pace of publications – which has reached an unprecedented frenzy during this pandemic period.

We note, however, that large publishers of journals have found a way to monetise articles rejected by their most selective titles by accepting them in new journals, often in ‘open access’ and therefore paid by the authors and their laboratories. Thus, articles rejected by prestigious journal A, but recycled in a less visible journal C from the same publisher, are still money makers, contributing more to the company’s economic profit than to the benefit of science.

Yves Gingras is scientific director of the Observatoire des sciences et des technologies and Canada Research Chair at the Université du Québec à Montréal (UQAM), Canada. E-mail: gingras.yves@uqam.ca. Mahdi Khelfaoui is associate professor at UQAM. E-mail: khelfaoui.mahdi@uqam.ca.