Kinder peer review

This is a guest post by , Executive Editor, Nature Partner Journals

Every day, thousands of scientists around the world donate their spare hours as peer reviewers to help colleagues (and competitors!) improve their work. But unkindness does exist too (as you can see from the emergence of Facebook groups such as …) and we all have a role to play in making it a kinder, more-productive process. As an editor, I have seen the full gamut of reviews, from unhelpful one-liners, through useful assessment of the work that highlights deficits and provides solutions to help the authors transform their paper, to unrealistic demands that go far beyond the scope of the paper under scrutiny.

There is a lot of comment out there on what makes a good scientific review, but what basic tenets of peer review could we agree to sign up to if we all wanted to make science a kinder place? Importantly, all parties involved in peer review need to remember that there are people behind the science. A publication is the outcome of hard work and time away from loved ones or much-loved leisure pursuits; for some, there are hopes, dreams and grants at stake. Fair, fast, thorough and impartial assessment is needed to ensure the wheels of research keep spinning.

Firstly, what can editors do? We can be transparent in our processes and keep authors informed. We can ensure we contact the best people to review a paper, and we can endeavour to provide a fast, fair decision, with guidance regarding the peer review reports and how authors might address the comments. We should ensure that we invite reviewers who represent the full spectrum of researchers. We should look for ways to support authors and referees in a continuously changing publishing landscape and to improve the peer review process by trialling new approaches that could help speed up peer review.

What can authors do? Please consider your readers, taking into account non-native English speakers, and ensure that your manuscript is written in clear and straightforward language and avoids scientific jargon. Please describe your research without any hype. Also, consider that readers may not be well versed in the methods you are using, and they likely do not have time to read up on previous research for context, so please include all essential details in your paper. When drafting display items, remember that some readers may be colour-blind or partially sighted. The data you created or used during the project should be clearly listed, and provided for review. Finally, please do complete the checklists provided by the journal; they are designed to make the editorial process go smoothly, to help identify issues early, and to make research reproducible – not as hurdles to overcome.

What can reviewers do? Please only agree to review papers that you are confident that you will be able to assess thoroughly and within the timeline requested. If you intend to ask an early career researcher to perform the review for you, please ask the journal to invite them directly so that they can start to build their own CV. If at any time you realize that you have a conflict of interest, please let the editor know and excuse yourself from acting as a referee. Please assess the paper you are looking at – not the person. Each new submission must be assessed on its own merits. If you are not sure what an insightful review report looks like, check out journals that publish or take part in that assess preprints.

Last but not least, thank you to all of those authors and reviewers I have worked with who have shown how supportive and constructive peer review can be. Papers shepherded through this process come out improved, more insightful and with clear direction for future research projects for the benefit of all.


  1. Pingback:
  2. For another perspective on the complex ecosystem of reviewers and reviewing…
    A clarion call to the community of current and potential journal reviewers
    Michael N. Liebman and Franco Marincola
    Journal of Translational Medicine 2018 16:200
    19 July 2018


  3. At the cutting edge of frontier science, matters are nebulous to all — authors, reviewers, and editors. There is a segment of authors that push papers/article through sheer eminence, creating big ideas that transfix science through assumptions and myths, e.g., cortical spreading depression in migraine research. Another segment of authors are roped in by the industry to carry out randomized controlled clinical trials to bludgeon in pharmaceuticals or devices, e.g, the Amplatzer device to close patent foramen ovale in migraine. Many authors revere the p value but abhor logic. Such researchers need to be rooted out of medicine as they are not in pursuit of the truth. What can kindness, a higher moral impulse, do here?


  4. Some journals reject manuscripts without any comments to the authors and this is not fair. The authors deserve at least some advice or commnets to their work.
    SalemY Mohamed
    Editor in chief
    African journal of gastroenterology and hepatology


  5. It is readily painful to authors for a Journal to receive a manuscript submitted online and leave it on Editor’s desk for 2-4 months unattended to. Such Journals would just wake up one day and reject the manuscript. Editors should endeavor to check submitted manuscripts within 1 to 2 weeks and act accordingly. If the manuscript is rejected in time, the authors would be free to make submission to another Journal. If the manuscript is suitable, then it should go to Associate Editor/review without waste of time. Every author appreciates quick actions on his or her manuscript. This is what I like about Editors of Elsevier Journals. They act on manuscript in time.


    1. Thank you for the comment. Indeed, science moves fast and a key part of kind peer review is to make sure that every editorial stage is as fast and fair as can reasonably be achieved.


  6. The best reviewer should have in mind “Please assess the paper you are looking at – not the person” reviewers should focus the paper data rather than finding relations to the authors sometimes we have to review a paper from any of our friends, so we must be honest in our job.


  7. Medicine is no longer the gentle, kind enterprise that it once was. Small-scale research efforts motivated PRIMARILY by the satisfaction of finding new knowledge through an uncompromising search for the truth is no longer the norm. Satisfaction has been replaced by glee. Volume of publication has far surpassed value. The RCT has ensured that multiple versions of the truth abound (Gupta, Expert Rev, Neurol., 2010). Trust has evaporated and been replaced by lust for fame and fortune. If you cannot trust your own doctor, who will you trust (BMJ)? If your own doctor recommends participation in a trial with dubious background or manufacturer-industry pressure, what are the options left to you? We have no system to control such fishing-expeditions. The Ethics Committee is an eyewash (Gupta, 2010). Statistics have kidnapped biology. Medicine has no checks and balances, as in a democracy all are meritorious and all must win prizes. However, medicine is not democratic as has been proved repeatedly. There is also no clone for Einstein. This spurious structure of medicine has to be torn down, not by gentle hands but by an intellect that goes beyond such puerile ventures. Such broad and capable minds are far and few (Alex Carrel — Nobel Laureate in, Man, The Unknown, 1959)


  8. What is the key role of Editors–a gentle role in medical publication?

    Writing successfully in medicine and neurology since 1989 — all the while swimming against the tide — I have a huge and UNIQUE chest of critical observations and critiques of the Editor-Reviewer combine.

    I will share only one with you, that was put forward by the sacked Editor of JAMA, Drummond:
    “The sole purpose of Editors is to keep egg off author’s faces”.

    If your logic or interpretation of the article or paper or previous established dogma, or, of your submitted data is superior, most Editors wrinkle their noses in disgust, unless the wind is blowing from Harvard/Oxford and allied institutions. Time and again, it has been proved that discovery is not democratic or confined to just a few countries.

    My article on Botulinum toxin was kept at the Editor’s desk for more than one year after all resubmissions and so-called corrections had been approved (Pain Medicine 2006; 7: 386-394).

    This is gentle?


    1. One role of editors, in my view, is to shepherd a manuscript from submission to publication. Including getting the most appropriate peer reviewer advice, assisting authors with peer review report interpretation, ensuring the manuscript contains the relevant details to be reproducible, assisting in identifying relevant data repositories, and making suggestions to ensure that the language is accessible. This should be completed with no consideration of where someone is based, or what stage of their career they are at.


  9. What is also very frustrating is the expectation of reviewers that the latest whizz bang method has to be used to answer a research question. For example, next generation sequencing seems to be becoming a requirement for evolutionary studies when other sequences, e.g. mitochondrial DNA, might be just as informative. It all depends on the question being asked. Not everyone can afford to use the latest techniques. Do these requirements by reviewers put all past research in doubt because those methods were not used? Also, and probably most importantly, it exacerbates the divide between researchers with lots of money, and those without. I would like to see reviewers be more realistic, and kinder, and realise that, even if the latest method hasn’t been used, the research question has still been answered adequately, and not place unreasonable burden on the authors trying to publish.


    1. I think that this is where reviewers can be mindful, but editors can and should also be providing guidance of what might be a nice extension to the paper, and what is required to establish the validity of the data in the current manuscript – these are two separate things


  10. It’s kind of laughable having a blogpost like this from a journal where you send your very best papers and in all likelihood get no reviews back at all!


  11. Is’nt Rebecca the only one in this thread using “can” and “should”? These hedge terms are effete expressions of a bygone era.
    Editors vary, like humans, in their approach to their task — which involves approaching the truth, which in turn has many and competing versions today. I have been fortunate to also experience the better sides of many editors, and have received genuine praise from such unexpected quarters.


    1. I am approving this comment partly in order to respond to it and to clarify our comment policy.

      We are a web-based magazine looking to change STEM culture by encouraging people to be more kind. Rebecca’s article is part of a series looking at ways in which aspects of the scientific method and scientific culture could be better – more thoughtful, more friendly, more kind.

      Thus terms like “can” and “should” are absolutely the terms to use. We are imagining a better STEM, and making suggestions as to how we can get there.

      Accusing an author of using “effete expressions of a bygone era” is not in line with our mission – it’s an unnecessary attack, on the author for where she works, not considering the article as intended in the context in which it appears.

      If you are unclear on our aims, please check out our Manifesto: /manifesto/

      Hopefully you will agree with us that the mission of considering STEM culture and trying to improve the way we interact with each other is a worthwhile one. We welcome comments from critical friends – but please, do try to be more kind.

      Yours –
      Hannah Dee


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. (  /  Change )

Twitter picture

You are commenting using your Twitter account. (  /  Change )

Facebook photo

You are commenting using your Facebook account. (  /  Change )

Connecting to %s