Guest Editorial: Elsevier General Counsel, Mark Seeley
By sticking with science and transparency, we can give assurance to society that our publishing processes are trustworthy
By Mark Seeley Posted on 17 September 2013
Ethics in society and business has many meanings — conducting one’s self in a way that respects and recognizes others and their contributions, but also the notion of compliance with our societally-sanctioned behavioral processes (including laws and regulations). Ethics in scientific and medical publishing takes these meanings a step further — by touching on the relationships within the scholarly communication process involving the journal, the relevant scientific or medical community that a particular journal serves, and society at large, publishing ethics encompasses a complex set of relationships which are mutually reinforcing. Scholarly journals have enormous prestige due to the stewardship that such journals have exhibited, over the decades, concerning the ‘record of science’ and the trust that their respective communities and society as a whole have in that record.
In one sense, publishing ethics allegations diminish that historical reputation and trust — climate change ‘deniers’ have used some controversies concerning undeclared potential conflicts of interest to suggest that the fundamental research is flawed, even though there is little evidence of substantial scientific disagreement on the broad question of the impact of industrial activities on climate. On the other hand, increasing the transparency and visibility over processes for managing publishing ethics allegations could — and should — shore up the journal’s reputation.
Society has highly unrealistic expectations of scientists, journal editors, and journals. The concept of peer review, for example, is often portrayed extremely simplistically as a ‘quality testing’ process —and consequently society expects that if an article has been through the peer-review process it should be completely sound in its calculations and methods. The reality is that peer review is about some fairly subjective points — a paper’s potential impact and importance and how the paper fits into the theoretical developments in the relevant discipline. It is also sometimes about identifying true ‘outliers’ in research results or methods, thus giving the journal and the editor an opportunity to review these in more depth. It is not a ‘Consumer Reports’ second laboratory duplicating the purported results of a particular paper for testing and quality assurance purposes. And yet, society is not wholly wrong to expect that the scholarly community — and thus the journal — should be better able to recognize fraudulent results or plagiarism and report on such violations as responsibly as possible.
The responsibility for correcting the scientific record when ethics allegations are confirmed falls largely on the journal editor and the publishing team, often with support from the relevant research funders, universities and institutions, and other investigators or peer reviewers. While Elsevier is adding a small team of publishing ethics experts, and while Elsevier legal team members are often involved in highly contentious or ‘legalistic’ matters, the expertise of the journal editor is fundamental to any significant investigation or consideration. The editor has the appropriate grounding in the particular discipline, an understanding of the relevant expertise of various research teams, and a general sense of which facts are more likely correct than not. We must always keep in mind that the standard of ‘proof’ that we look for is a general ‘more likely than not’ standard, not a deeply rigorous, criminally-oriented, ‘beyond a shadow of doubt’ standard. One reason for this standard level is that we anticipate that the relevant scholarly community, as an informed reader/consumer of journal content, will come to its own conclusions as to the merits and relevance of particular allegations and claims, assuming the requisite transparency and disclosure.
Law is the profession of skeptics, and a skeptical point of view is often useful in judging the merits of competing narratives in a publishing ethics dispute. Law is supposed to teach us to be logical, to think through the alternatives and contrary points of view, and to disregard emotion and ‘threats’. There are sometimes vague and sometimes very specific threats that are made by complainants or subjects of inquiries about resorting to legal process and formal legal complaints. These are generally quite silly. The few courts that have been asked to opine about scholarly publishing complaints have generally been respectful of the scientific process and charmingly denigrating about the ability of the legal process to improve on the underlying scientific investigation or conclusion. But my team is prepared to help when complaints like this are made or when it is otherwise deemed useful by editors or publishing team members. We are happy to help, and we are quite passionate about publishing ethics issues and process.
For me personally, I have been involved in publishing ethics policy discussions, the drafting of policy and procedural documents, individual investigations of a ‘legalistic’ nature, and service on our formal ‘retractions panel’ for more than ten years now (along with my UK-based colleague, Senior Vice President of Research & Academic Relations, Mayur Amin). I have enjoyed my discussions with editors and our publishing team members on these issues, and I like to think I have helped to contribute to good policy-setting and reasonably professional retraction processes. I’m also fairly opinionated, so I wanted to conclude this introduction with some general comments and observations.
One can make the argument that the level of ethics issues has not significantly increased, but is simply more visible now. However, I think the better view — one more consistent with the evidence on the number of retractions — is that we are seeing an actual rise in volume. The number of formal retractions as recorded on Elsevier’s ScienceDirect platform has more than doubled between 2004 and today — in fact, for 2013 it looks as if we will have close to 200 retractions, which would be five times the number we had in 2004. I believe this is due to the ramp up of the pressure to publish and occasionally the pressure to take short-cuts — and I believe this pressure is increasing across the board, including in the rapidly-developing countries of the world where scientific research is exploding. As vehicles for investigation and comparisons, I applaud the efforts of COPE (the Committee on Publication Ethics) and CrossCheck (both of which you will hear more about in Part II of this Ethics Special), but would note that they are only vehicles and not a substitute for editorial judgment.
Researchers should be allowed the benefit of the doubt, particularly at an early point in their career, and we should accept that researchers will sometimes make mistakes at this stage which they can then learn from. This is why I am leery of the notion of ‘blacklisting’ an author or research group. I do not believe that authors should be given carte blanche when it comes to ethics violations — simply that we should be careful not to rush to judgment and be careful in recommending the appropriate sanction for the relevant degree of ethics violation.
I think that not all misconduct is equal — to me outright fraud is the most dangerous and has the most impact on the community (as it may cut off otherwise promising research areas), and fraud allegations are relatively rare. I would distinguish between plagiarism, which involves taking credit for someone else’s research efforts, as opposed to merely copying a stray paragraph or two — the latter is certainly wrong and deserves some form of censure but the former is more inherently improper as a destabilization of the research environment. I think arguments about authorship, and particularly the idea of one author as opposed to all co-authors being identified in a retraction notice as the ‘culprit’, are unnecessary and ultimately do not advance science. Of course, I accept that a co-author who was an equal participant in a project and a paper deserves recognition for their contribution.
Elsevier has had to address some publication ethics issues over the past several years, but I believe we have addressed them head-on and pro-actively. Last year, there was some discussion about the ‘faking’ of peer-reviewer identities in our article submission system, which we acknowledged and dealt with by improving our system (unfortunately identity ‘theft’ in this sense is difficult to prevent entirely except by improving on personal security systems in areas like passwords). We have also had public controversies about a now-retired editor who accepted many of his own authored papers for his journal, and controversies about genetically modified foods and the raising of children by same-sex couples. In all of these controversies we have worked hard to achieve the ‘right’ resolution — first by emphasizing science (if appropriate scientific and publishing processes have been followed, then our view is that we should stick with the result, even if it is not ‘politically correct’) and second by emphasizing transparency and disclosure. Again, we trust the relevant scientific community to put things into context and judge things on their scientific merits.
By sticking with science and transparency, we can give assurance to society that our publishing processes are trustworthy — not that they are perfect but that they can be relied on. Science publishing, however, is not headline-oriented, short-term journalism — it is about the long-term process of building on discoveries and theories.
I hope you enjoy this special edition as I know I will!
Senior Vice President & General Counsel, Elsevier
Chair of the Copyright and Legal Affairs Committee, STM (International Association of Scientific, Technical & Medical Publishers)
David L. Keller, M.D. says: September 20, 2013 at 12:37 pm
The main problem is not minor infractions by authors, but rather the fact that large studies are often funded by the manufacturer of the product being tested, and the fact that the investigators accept payments as consultants. Recent suspicious results include the downplaying of adverse effects in the recent DBS study in NEJM, as pointed out in letters to the editor.
David L. Keller, M.D. says: September 20, 2013 at 12:47 pm
I once submitted letters to two different Elsevier journals making similar comments regarding the review articles they both published on the same topic around the same time. I was pointing out an oversight which could affect patient safety. One journal editor decided there was no problem with dual publication and printed my letter. The other editor said it was a violation of publishing standards and withdrew acceptance of my letter. Later, they printed a revised version which added additional material. What is the standard for timely safety warnings in multiple letters to different journals?
Mark Seeley says: October 15, 2013 at 11:34 am
It is always difficult to comment on specific circumstances without knowing all the circumstances, but in general I agree that the principle of “no double publication” wouldn’t normally be relevant for letters or other communications, particularly when the issue deals with potential patient safety issues. Useful to know, however, that at least the one letter was published!
Tsung-Chih Tsai says: October 27, 2013 at 9:41 am
This is indeed very necessary
etseq says: January 21, 2014 at 4:48 am
I think it is very telling that you used the right wing phrase “politically correct” when defending the editorial misconduct that allowed Mark Regnerus to use your company to publish a political hit job on gay parents. You must be aware of the internal audit by Professor Sherkat that detailed the conflicts of interest among the referees and the transparent methodology used to generate the highly inflammatory findings, specifically the allegations that gays and lesbains molest their own children. Emails released from the UT Austin revealed a coordinated campaign between Regnerus, Brad Wilcox (a board member of your journal, a paper referee, a consultant on the study and an official with the conservative think tank that funded the study) and the Witherspoon Foundation to produce a negative study for the purposes of influencing the supreme court.
Your company has attempted to intervene in la Florida lawsuit seeking documents under that state’s open records law. This is not about the sanctity of peer review – its about your attempt to cover up the ethical misconduct that allowed Regnerus and his political allies to manufacture fake science to justify their political homophobia and slander gay parents in the process,
Thankfully, the academic community recognized that the paper was a fraud and the AMA, ASA, and APA all filed briefs to explain to the court what your editor and his flawed peer review failed to recognize.
Tom Reller, Vice President Global Corporate Relations says: January 23, 2014 at 3:56 pm
Hi Jimmy, thank you for taking time to comment on Elsevier Connect. I’m happy to elaborate a bit on this issue, as I’m afraid you’re not describing the situation correctly. What’s at issue here is that Elsevier respects public records requirements and public institutions’ compliance with properly drawn record requests, and we are always willing to seek reasonable compromises when initial requests are overly broad and implicate sensitive private information.
In this instance, Mr. Becker has broadly requested every email sent or received by Professor Wright, regardless of the topic or the possible private nature of the communications. The University of Central Florida has objected to this request on the grounds that much of the material requested does not qualify as public documents within the meaning of Florida law. We have moved to intervene because Mr. Becker’s request seeks the identity of all peer reviewers that have reviewed articles for Journal of Social Science Research. To our knowledge, every court that has ever addressed the topic has agreed that the identity of peer reviewers should be kept confidential, for the good of the scientific process as a whole, and Elsevier seeks to have this rule applied here.
The AMA, ASA, and APA have not filed briefs in this matter and we would be surprised if those organizations, whose members frequently act as peer reviewers, would take a different view of this issue. We believe that the identity of peer reviewers for each of our journals should also be withheld because it is proprietary information. We would not object to Mr. Becker receiving more narrowly tailored information relating to this matter, if the information falls within State disclosure requirements, so long as UCF properly withholds or redacts information that is specifically exempt from the State disclosure laws to protect private parties.
I hope this is helpful information. Thank you.