The Opinion Piece: How Blind Should Peer Review Be?

In the first of The Opinion Piece series, editorial intern Joe Davies explores the issues involved in anonymous peer review, and asks: how anonymous should it be?

The peer review process is the back-bone of the scientific method. If an author wants to have their research paper published it must go through a process by which it is critiqued by other scientists to test the validity of the work. This can range from simply checking that the ideas involved make sense to repeating an experiment in full to check whether the results match up.

One of the main tenets of peer review has traditionally been that it’s anonymous in one direction (known as ‘single-blind’); the author does not know who is writing the review, even if the reviewer knows who the authors are. The idea of the anonymous review is ingrained in the academic landscape, so it may come as a surprise to many to learn that this idea could be changing. The question is: should peer review really be anonymous?

A Brief History of Peer Review

The act of peer review (at least in science) is credited to Henry Oldenburg, a German theologian, known as a diplomat and natural philosopher. He was alive during the 17^th century and acted as the very first secretary for the Royal Society after its foundation in 1660. As a popular diplomat he had a web of contacts all over Europe, including: Robert Boyle (one of the founders of modern chemistry), Samuel Hartlib (British-German polymath) and John Milton (poet known for Paradise Lost).

In the role of secretary, and the founding-editor of the journal Philosophical Transactions of the Royal Society, he used these extensive contacts and began sending submitted manuscripts to experts all over the world in order to get their opinion on them before publishing. This sending of manuscripts to be vetted before publishing began the process of modern peer review and of the scientific journal.

What’s the Problem?

One big contention point is the anonymity of peer review. Proponents of anonymity state that it places everyone on a level playing field, as not knowing who is reviewing your paper makes everyone’s opinion hold the same weight. This is a boon for younger reviewers (who may also do a better job of it¹) as it protects them from retaliation to a negative review from older, better established colleagues, in the form of denied grant applications and future jobs. In fact when Nature contacted academics for a comment on this, many said they wouldn’t review anymore if forced to sign their names² for fear of such retaliation.

In contrast, lack of anonymity may lead to favouritism, and adversely affect the quality of reviews. A reviewer that has to reveal their identity may give preferential treatment to prominent authors or organizations in order to `stay on their good side’. This can make reviews more timid, leading to a review process that doesn’t do much to combat sloppy research or wild claims. More thorough reviews mean that only the best version of a paper is let through.

Of course, there are downsides to anonymity and non-timid reviews. As much as we would like to think otherwise, some researchers are judged not only on their academic ability but also skin colour or gender. Bias against certain groups of people, especially women and POC (People Of Colour), has always been a problem in science³ and though the situation has started to improve⁴, it is still an issue. Single-blind peer review means that bias can effectively continue because the reviewer can feel safe in the knowledge that the author won’t know who they are. A study in 2015 by Professor Joan Williams found that 93% of white women and 100% of women of colour who were surveyed had experienced some form of work-place bias⁵.

In addition, the ramifications of leaving a nasty or too-personal review aren’t always far reaching. If a reviewer were to leave a clearly biased or even overtly prejudiced comment journals can find it difficult to impose the sanctions that the same rhetoric would get in other areas. This is because peer review is a largely voluntary process, meaning the consequences for inappropriate behaviour aren’t as far reaching as they would be in a paid environment. Even if a particular journal chooses not to use them anymore, there’s nothing stopping them from reviewing for another journal and doing the same thing. Removing anonymity could solve this issue by making the reviewer more exposed and likely to think twice before saying something overly partisan.

Finally, anonymous peer review helps to hide the fraudulent practices of predatory publishers. Predatory publishers feign peer review in order to do no work but still charge a fee to the author. The peer review process is anonymous so these publishers can, and often do, pretend that a paper underwent peer review and then, when prompted to reveal how they did it, can hide behind the anonymity defence. Non-anonymous reviews would let individuals, who may have been duped by predatory publishers, check the credentials of the person reviewing. Of course this can then open the doors for authors to complain that a reviewer is biased or inexperienced even when they aren’t.

The Opinion Piece

Some organizations are now talking about rolling back anonymity. Most notably, the British Medical Journal (BMJ) has completely removed it⁶, and their current model has reviewers signing their comments by name. Their reasons for doing so, they say, are ethical with the editor, Richard Smith stating:

“A court with an unidentified judge makes us think immediately of totalitarian states and the world of Franz Kafka” ⁷

To me, this is a bit of an over-exaggeration. Peer review is less about sentencing a paper to be good or bad based solely on a `jury of your peers’ and more an advisory process to let an editor know what papers would work best in their journal, and to help the authors produce the best paper possible via constructive feedback.

In my opinion, we should double-down on anonymity and make the author (and institution they belong to, if any) of a paper anonymous as well. That way we can cut down on the bias that comes from reviewers without opening the doors to timid reviews or nepotism. Some individuals may be able to figure out who they are reviewing from particular style choices or research topics, but I doubt this would be commonplace.

Anonymity in peer review is something that should stay. Science research is supposed to be judged purely on the paper in question and the personality of the individual shouldn’t matter. It boils down to the merit of the research, not the researcher’s merit.