A version of this article in French exists : L’édition scientifique : son modèle, ses scandales.
The first two academic scientific journals made their appearance in 1665: Le Journal des Sçavans and Philosophical Transactions of the Royal Society of London. Their aim was to publish scientific articles to encourage their dissemination and develop the long-term archiving of scientific results. In 2009, there were about 24,000 peer-reviewed journals with about 1 million contributing authors. On average, 1.5 million publications are read each year by 10-15 million readers in over 10,000 institutions .Thesenumbers increase by 6-7% each year. Scientific publishing houses were formed in order to regulate this proliferation of publications. Their primary role is to publish scientific results and maintain the quality of publications. Over the last few years though, a number of scandals have sullied the reputation of the large publishing houses, adding to regular criticisms of their functioning (business plan, peer review, price, etc.)
The three most important of these publishers are Thomson Reuters, including Thomson Scientific, Reed Elsevier, including Elsevier, the branch that publishes scientific and medical reviews, and Wolters Kluwer, including Springer. These three alone account for 90% of the market. Other practices, mostly linked to the development of the Internet, have emerged alongside publishing houses. Open access and self-archiving are representative of the break between the monopoly established by certain publishing houses and the complete openness of the web.
It’s (publishing) house policy
The commercial purpose of scientific publishing houses is to publish scientific results, generating large profits in the process. The authors of these results, whether involved in academic or private research, aim not to benefit financially from these publications, but to spread their results as widely as possible in order to develop collaborations and increase their renown. From the point of view of institutions that produce scientific results, the goal of publishing these results is not to make a profit. On the contrary, although submitting an article to certain journals is still free, the fees for publishing in several prestigious journals are such that they impose a high price on institutions to be able to publish in major journals.
Let us take the prestigious medical journal PNAS (Proceedings of the National Academy of Sciences) as an example. It publishes articles varying between 6 and 10 pages in length:
PNAS Plus articles will cost $195 per research article page, with no additional charges for the author summary, color figures, […] extract from editorial policies of PNAS."
Thus, a number of journals require an article submission cost of over 1000 dollars. Other journals state that submitting an article to their publication is free. Nevertheless, authors and their institutions are often asked to pay ‘optional’ fees to obtain services that are, in fact, anything but optional. Including a colour figure in an article submitted to Nature is one example of this:
A contribution towards the total cost of reproduction of colour figures is requested. We currently charge £757.05 for the first colour figure and £270.38 for each additional figure. Inability to pay this charge will not prevent publication of colour figures judged essential by the editors, but this must be agreed with the editor prior to acceptance."
However, scientific articles with neither graphics nor figures are rare. Written scientific communication is based on two steps: publishing the results as we have just said, and reading articles that spread the results of others. So, belonging to an institution with subscriptions to a certain number of well-considered journals is essential to be able to access articles. Institutions, of course, have a limited budget to put towards these subscriptions.
Subscription fees for the most influential journals are high, and they are still rising, despite the progressive move from printed to electronic versions and the improvement of machine-driven automatic archiving. Thus, the price of scientific journals hasincreased considerably (4% to 5% per year) despite the economic crisis, something that, when combined with recent severe budget cuts, is putting academic libraries in a difficult situation .
Furthermore, publishing copyright policy demands that authors transfer their rights to their publisher. These rights authorize publishers to publish the text at their own expense and to be the recipients of the income generated by the text. After transferring their rights, authors have neither privileged access to the published online version, nor the right to use the text in any other form except that in which it is published. In the same way, authors of quality scientific books – the fruit of the intense labor of several scientific specialists in the field – are only lucky enough to receive one, or perhaps two copies of the work that they themselves have written. These books, often priced in the hundreds of euros, are thus an offering of time and energy made by researchers, in the name of Science; publishers are, in fact, the intermediaries and, at the same time, the bankers.
However, some publishers (over 60% of the publishers referenced in SHERPA RoMEO) authorize researchers to deposit a copy of their articles into an open archive after a period of embargo of varying length.
These publishing policies are quite common to most scientific publishing groups. Exceptions to this include open access publishing journals that independently developed a model different t o that described up to this point. It is true that scientific publishing works because it is supported by researchers who work for free, pay to publish their work, but also – and this is surprising to someone who has not been caught in this vicious cycle – pay to read others’ publications. The logic behind this can be found, though, in the history of scientific publishing and science itself. Due to the hyperspecialized nature of scientific disciplines, scientific publishing cannot be entrusted to a non-specialist in the same way as the publication of a novel, for example. For this reason, assessing the scientific quality of an article’s content in scientific publishing is based on peer-review; this system requires that the work of reviewing, evaluating and making suggestions or correcting is undertaken by researchers who are specialized in the scientific field of the article. The trust placed by a reader in the results published is based on the free work of researchers: the same people who pay to publish their own work.
During this process, besides being the intermediary between authors and referees, the publisher is in charge of making minor changes on the language level, the quality and size of figures, and the final layout. A close reading of the publication guidelines shows the extremely restrictive character of the submission process: size of figures, text format and vocabulary to be used… All of this is highly regulated and it is the responsibility of authors to follow all the rules for every article they submit. Publishing an article in a highly prestigious journal is a job in itself. Not only is it very expensive and time-consuming, it also and above all limits communication to studies that are the most “in fashion”.
The financial plan of some of the most influential scientific journals is largely based on personal and institutional subscriptions. The practice of grouping sales, referred to as “packaging” or “bundling”, typically combines one or two highly influential journals with a number of less influential journals whose subject has little or no relation to the interests of the subscriber (these are called “Big Deals”).
[...] Each publisher has one or several model contracts that change over time. These contracts generally include confidentiality clauses that make their systematic analysis impossible.
For the sake of simplification we will refer to a “Big deal” as a multi-year contract in which a library undertakes to increase the total amount it sets aside for the purchase of a publisher’s journals. In return, the publisher undertakes to give electronic access to a larger collection of journals than are already available through the library’s print subscription. Electronic journals are thus sold as a bundle. Furthermore, the fees paid by each library differ depending on the amount of money previously spent on print subscriptions."
It is common then for institutions to subscribe to a large number of relatively irrelevant journals, as the package system cannot be worked around.
Beyond a policy of high pricing and removal of the author’s copyright, slowness is the most problematic aspect of the publication process, which lasts 10 months, in some cases, or even more. This delay can penalize a scientific career given that, in France for example, several publications are required to be able to defend a doctoral thesis that must be completed in 3 years (the time allowed is becoming more and more strict). During the submission process, and several months after the article is first sent in, the author and the referee exchange suggestions and corrections through the publisher who acts as an intermediary. This process leads to either the outright rejection of the article or its acceptance for publication with or without modifications. At this stage, after several months during which corrections were made as referees asked for further experiments, authors must permanently and irrevocably (provided that the agreements’ conditions of validity are observed) relinquish their copyrights over the article to the publisher .
The long-term archiving of publications managed by publishing houses works on the basis of a fully capitalistic system. The reliability of such a system is questioned even though digital archiving is the main argument used to justify the high prices of journals and access to journal platforms.
Scandals and scientific publishing
Elsevier, among all scientific publishers, is the publishing house that has caused the largest number of scandals. In June 2007, the publishers of the medical journal “The Lancet” (owned by Reed Elsevier) rose up against their parent-publishing house in an ethical editorial accusing them of promoting the sale of weapons. Though legal, this activity did not seem compatible with the aspirations of the journal to spread medical knowledge, according to those involved. In addition, The Lancet was highly implicated, at the time, in a controversy claiming that the number of deaths in Iraq was higher than what had publicly been announced. Several petitions circulated and demonstrations brought together a number of researchers, doctors and professionals who published articles in these journals to denounce Elsevier’s conflict of interest. In the end, the publisher put an end to its activities related to weapon sales in May 2007.
Editorial of The Lancet (2007) : Reed-Elsevier’s hypocrisy in selling arms and health [issuu layout=http%3A%2F%2Fskin.issuu.com%2Fv%2Flight%2Flayout.xml showflipbtn=true documentid=110705153805-803489bfbeaa45e08ea28941e0b89580 docname=thelancet username=yolobia loadinginfotext=R%20e%20e%20d%20-%20E%20l%20s%20e%20v%20i%20e%20r%20%E2%80%99%20s%20%20h%20y%20p%20o%20c%20r%20i%20s%20y%20%20i%20ns%20e%20l%20l%20i%20n%20g%20%20a%20r%20m%20s%20%20a%20n%20d%20%20h%20e%20a%20l%20t%20h showhtmllink=false tag=editor width=420 height=272 unit=px]
Conflicts of interest are particularly troubling in the fields of health and pharmaceuticals. This was underlined by another accusation laid against Elsevier in 2009 that blamed the group for publishing six false medical journals, sponsored, in fact, by large pharmaceutical groups. Their goal was to publish papers that resembled scientific articles while promoting certain therapeutic substances. Other cases were to follow. The massive and unjustified publication of certain influential authors is one example. Another example is the case of ghost writers publishing articles in medical journals on behalf of the pharmaceutical industry. Other cases of false journals and article duplication also came to light. Basically, scientific publishing is partially corrupt, under the cover of influential journals read by thousands of doctors and researchers.
Despite the peer-review system, devised to guarantee the quality of published articles, the article selection process is not exempt from errors. Towards the end of 2010 for example, a group of researchers announced the discovery of a bacteria capable of using arsenic (a toxic substance) for phosphate in its DNA. During its long awaited and internationally relayed press conference, NASA and the group of scientists concerned wrongly announced the possibility of a new form of (extra-terrestrial) life, very different to those we are currently aware of. After a significant media buzz and only two days after the article appeared in Science, this result was refuted by researchers who pointed out – on their blogs and via twitter – the unreliable methods and rushed conclusions of the study presented . NASA, the authors of the article and the publisher dismissed the critics on the grounds that the discussion was not legitimate, and did not conform to the peer-review evaluation process. The unanimity of scientist-bloggers/journalists finally succeeded in getting their message out and Science published comments/discussions relating to the arsenic bacteria. This episode seriously called into question the process of peer-reviewing for this article. No less than nine comments were published on 3 June 2011:
NEWS & ANALYSIS MICROBIOLOGY
Concerns About Arsenic-Laden Bacterium Aired
The debate that erupted 5 months ago over whether a bacterium could thrive on arsenic, even incorporating the element in its DNA, is finally being aired in the scientific literature rather than on blogs. Originally published online by Science on 2 December 2010, the paper describing this potentially unprecedented microbe appears on page 1163of this week’s issue. Eight Technical Comments, released online last week, formally raise many of the criticisms that were quickly hurled when the study was published online and publicized in a NASA briefing. The scientific exchange is unlikely to be the final word on the contentious issue, all agree, especially because there have yet been no independent studies of the microbe at the center of this spat. Elizabeth Pennisi, Science 3 June 2011
Technical comments on a bacterium that can grow by using arsenic instead of phosphorus
Wolfe-Simon et al. […] reported that bacterium GFAJ-1 can grow by using arsenic instead of phosphorus. However, the presence of contaminating phosphate in the growth medium, as well as the omission of important DNA purification steps, cast doubt on the authors’ conclusion that arsenic can substitute for phosphorus in the nucleic acids of this organism. Rosemary J. Redfield, Science 3 June 2011"
Universities boycott scientific publishers
The policy of “package” journals and the high fees of institutional subscriptions have both led to several boycotts by prominent universities over the last few years. The prestigious Max-Planck institute cancelled its subscription to Springer in 2007, because no agreement on fees and services could be reached. Then, at the beginning of 2010, the Universitiy of California started a boycott of journals published by the Nature Publishing Group (NPG) – to which the Californian universities supply a large fraction of the articles – accusing them of increasing their fees by more than 400% at the height of the economic crisis. Recently, at the end of 2010, it was the Pierre and Marie Curie University’s turn to cancel its contract with Elsevier, presumably for similar reasons.
Among the many scandals that affected scientific publishing, we have decided here to consider only those with the greatest repercussions. University boycotts of publishers bring to the fore the excesses of a lucrative system, subject to the monopoly of a few influential publishers. Scientists – humans above all else – are the victims, but also the guarantors of this empirical system
Evaluating scientific research
A recent function of the bibliography is to evaluate scientists by their published works. This practice has, unfortunately, led scientific publishing astray. Today, along with its role of spreading scientific results – something which, in itself, amounts only to communication – scientific publishing has become a key factor in evaluating scientific research through the number of articles published and the impact they have.
Scientific journals are classified in order of content importance – thanks largely to the Impact Factor (IF) and the eigenfactor – so that readers are not drowned in the abundance of journals, but rather, can be easily directed to the highest impact journals. These two factors are calculated by sub-branches of Thomson Reuters and include numerous databases of scientific articles, the best known of which are the Web of Science® for Thomson and Scopus® for Elsevier.
"The impact factor is the relationship between the number of citations received by a periodical over two years and the number of articles published by this periodical over the same period.
Journal (example): EMBO Journal
Citations in 2005 of articles published in: 2004 – 4314 in: 2003 – 6624 2004+2003: 10938
Number of articles published in: 2004 – 462 in: 2003 – 626 2004+2003: 1088
Impact Factor: Citations of recent articles/Number of recent articles = 10.053 = IF"
This classification of journals allows the ranking of scientists’ studies on the basis of the journal in which they will be published. This is a simple way of “marking” scientists along with the Hirsch index or ‘h-index’ that quantifies the total number of publications by an individual and the level of citations of these publications (referred to as authors’ ‘citation reports’ in the ISI Web of Science only accessible to subscribers).
However, these evaluation indices are often criticized for their lack of transparency, their irreproducible nature and the classifying criteria used that privileges quantity over quality. The integrity of these factors has already been called into question and the classification of journals has recently been abolished in Australia:
This approach [removing global evaluation factors, IF etc.] will allow experts to make judgements about the quality of journals in the context of each discipline. Professor Sheil
We should also note that the highly controversial ‘Shanghai classification’ evaluates the quality of universities, especially in relation to the number of articles published in the journals ‘Nature’ and ‘Science’.
Using the bibliography in order to evaluate research could be considered legitimate, but it has introduced a change in publishing practices. Results are sometimes published prematurely rather than worked on carefully. The quest to publish in ‘Nature’ and ‘Science’ sometimes entails a significant loss of time when a submitted article has difficulty passing the barrier of multiple referees, and other articles are occasionally wrongly published and later refuted by more rigorous studies. As a consequence, some have made reference to this deviation of the publication system with the phrase “Publish or Perish” – a phrase that refers to the impact of publication on the evolution of a scientist’s career and promotional prospects. At present, then, obtaining a position at the CNRS (French National Center for Scientific Research) requires a minimum number of articles published, some of which must be in highly influential journals. As we have seen in the case of the arsenic bacteria, this leads to important data occasionally being under-reviewed.
To conclude, the criteria used to evaluate the impact of publications and research appear to be criteria of visibility more than quality. The competition between scientists is probably one reason for the limited use of social practices in data sharing. Web 2.0 does not foster a competitive spirit among its members but favors instead data sharing initiatives, consultation and discussion. Scientific communication undertaken by traditional publishers is a pillar of this system, which has been tarnished several times over the last few years by moral and ethical considerations. Doubts have emerged about the integrity of the system and critics are questioning its financing model.
New and still largely peripheral practices in the dissemination of scientific publications, and ‘open access’ in particular, offer an interesting and democratic alternative that would favor both access to publications and the visibility of scientists, while still offering quality content. Freeing itself from the system imposed upon it by publishers, scientific communication would be in a position to rethink its system, for example, by allowing authors to take back into their ownership the publication of their results using more flexible formats.
 Mabe MA (2009): Scholarly Publishing. European Review 17(1): 3-22  Aspects juridiques de la publication scientifique : Guide pratique à l’attention des membres de la communauté universitaire, Laurence Thys, (2009)http://orbi.ulg.ac.be/bitstream/2268/4277/1/Guide%20juridique%20CIUF%208.pdf  INTOXICATION MÉDIATIQUE À L’ARSENIC, 2011, http://www.knowtex.com/nav/intoxication-mediatique-a-l-arsenic_18280 En savoir plus : 1) L’évolution de la pratique des réseaux sociaux en science, L. Bianchini, MyScienceWork 2011, http://blog.mysciencework.com/2011/04/26/l%E2%80%99evolution-de-la-pratique-des-reseaux-sociaux-professionnels/ 2) L’open access : vers une nouvelle pratique de la communication scientifique, L. Bianchini, MyScienceWork 2011, http://blog.mysciencework.com/2011/07/06/l%e2%80%99open-access-vers-une-nouvelle-pratique-de-la-communication-scientifique/ 3) Periodicals Price Survey 2011 | Under Pressure, Times Are Changing, S. Bosch, K. Henderson, H. Klusendorf, 2011,http://www.libraryjournal.com/lj/ljinprintcurrentissue/890009-403/periodicals_price_survey_2011_.html.csp 4) Elsevier a boycotter?, Owni, http://ownisciences.com/2010/11/02/elsevier-un-editeur-scientifique-a-boycotter/ 5) Conclusions du Conseil sur l'information scientifique à l'ère numérique: accès, diffusion et conservation, 2007,http://www.consilium.europa.eu/ueDocs/cms_Data/docs/pressData/fr/intm/97240.pdf 6) Escape from the impact factor, P. Campbell, Nature, 2008,http://www.knowtex.com/nav/escape-from-the-impact-factor_2847 7) La liste Sherpa des éditeurs scientifiques : http://www.sherpa.ac.uk/romeo/browse.php 8) LES GHOSTWRITERS DES SOCIÉTÉS PHARMACEUTIQUES, OWNI, 2011, http://owni.fr/2011/06/10/les-ghostwriters-des-societes-pharmaceutiques/