The Ethics of Peer Review: History (1/5)
(May 10th, 2016) Peer review is at the heart of research communication. However, the process is far from perfect and many problems have been identified. Among them, ethical questions about the responsible conduct of reviewers who can wield considerable power behind a cloak of anonymity.
In an attempt to re-introduce ethical norms to scientific research, it has been suggested that researchers should adhere to a scientific equivalent of the Hippocratic Oath (discussed in LT 1/2016, pg 18-23). Such an oath would reinforce the principles of good conduct among scientists, in their work, their relationships towards other scientists, and their responsibilities towards society as a whole. Ethical codes also tend to define what is considered to be bad or unacceptable behaviour.
In addition to a general science oath, some researchers have identified specific areas of science where they say a well-defined oath could directly improve research practices by clearly stating key ethical principles. Peer review has provoked particular attention. The publication of research articles is at the heart of scientific research. It records what has been done, who did it, what their data looked like, and how this has been interpreted in the context of other research findings. Normally, authors will send their research manuscripts to journals and those that meet the required standards are accepted for publication.
However, the decision whether to accept an article is based on a reviewing process - the ‘peer review’ system. This was first adopted in the 20th century and has now become the norm for many journals. It involves an editor sending out a paper to a few experts in the field, who then provide comments for the paper’s authors. Although the reviewers can generally see who the authors are, they themselves remain anonymous to the author, and only the editor knows everyone’s identity. It is at this point that many problems can occur. Variability in how reviewers decide what is worthy of publication, where and when, causes considerable friction between authors, editors, and reviewers. Delays may set back research projects, affect ongoing research funding, and damage career prospects.
Scientific publishing has been around since the 17th century, but formal peer reviewing of submitted articles by external academics is relatively new. For example, the journals Science and the Journal of the American Medical Association (JAMA) only introduced formal peer review in the 1940s, and Nature didn’t introduce it until 1967. This is largely because science was still a relatively small enterprise until the Second World War. Many science journals were led by editors who could manage everything – “There were often more journals than scientific and medical papers to publish; the last thing needed was a process for weeding out articles”.
Yet, as science became more specialised, some science journals began to direct papers to distinguished experts who would serve on affiliated editorial boards. After the war, there was a huge expansion of science funding that resulted in a “deluge” of research manuscripts whose increasing specialisation made it difficult for even editorial boards to manage. The peer review system developed to meet this need. Journal editors began to seek out experts capable of commenting on specialised manuscripts - not only researchers in the same general field, but researchers familiar with the specific techniques and even laboratory materials described in the papers under consideration.
Overall, reviewers and editors have become “gatekeepers” in scientific publishing. By setting the acceptable standards for research publications, they decide what is allowed to pass through. In this way, it has been argued they serve to eliminate the most uninteresting or least worthy articles, and save the research community time and money.
In general, peer review follows a common path: researchers perform their experiments, obtain and analyse their results, write a manuscript to record their findings, and submit it to a journal where it will be considered for publication. If the journal editor decides to consider the manuscript further, it will be sent out for reviewing to 2 or 3 qualified researchers. This is where peer review comes into operation - the reviewers are presumed to be sufficiently competent to make a fair and objective evaluation of the manuscript's scientific content. Based on their reading, they should arrive at one of three judgements: 1. Is it good enough to be published as it is? 2. Should it be refused? or 3. Should the authors be given a second chance and told to improve their manuscript, either by modifying the text or by adding new research results (some of which may require new and time-consuming experimentation)?
Another key feature of the peer review system is that the author cannot know the identity of the two or three reviewers who are evaluating their manuscripts. The argument usually advanced for maintaining the anonymity of peer reviewers is that it allows the reviewer to make truly objective evaluations without any fear of reprisals or conflict should the authors object to the comments made. Unfortunately, there are many examples of the system’s abuse and dysfunction (discussed in Part 2).