Writing referee reports – the nurturing viewpoint

Uri Alon Nov 26 2008

As I write this, I'm recovering from yet another aggressively written referee report. I am grateful for useful criticism that improves my paper. I suffer from needlessly aggressive ways of stating criticism.
Aggressive language for example, is "The introduction is weak and misleading- it ignores previous work by X " as opposed to saying the same constructively "The introduction can be improved by citing previous work by X. This work is relevant because … ".

At present, referees receive no coaching from journals on how to give criticism.
This coaching is essential, because referees are put in an emotionally challenging position: given authority as experts, and at the same time anonymity. Unfortunately, we are not taught the basic psychological principle that authority plus anonymity tends to give rise to aggression. A cartoon of this well supported fact is the Milgram experiments, where people are told to administer electric shocks to ….[also Zimbardo experiments on prison]. In fact, if we pretend that we can ignore these basic psychological effects, we compromise the "objectivity" we seek in refereeing.

In the nurturing paradigm, we write referee reports to help the authors reach the full potential of their paper. We thus:
  1.  First look at the half full cup- the strengths of the paper. Avoid the bias in scientific education for over-criticism.
  2. Choose among the weaknesses only those truly relevant, not suggesting the myriad of tangential additional experiments that will bloat the paper.
  3. Write comments in non-violent language – how the paper can be improved.
  4. Remember the authors have been thinking about the paper for two years or more, whereas we have only a brief while to ponder it.

One can recommend rejection or acceptance, doing onto the authors the same as you would have them do unto you.
Why is non-aggressive language important? First, it lowers the barriers for authors to understand your comments, and eventually to accept them. Second, non-aggressive language helps to break the cycle of abuse: at present, aggression is passed along the circuit from scientist to scientist (treat 'em as I was treated). I find that aggressive referee reports are one of the main reasons that sensitive students can become fed up with science- they require quite a bit of healing after particularly harsh reports.

In writing nurturing referee reports, self knowledge is very important, because one must recognize traps which raise negative emotions and can lead to wrath on the paper. Common traps are: (a) I wasn’t cited - leading me to feel invisible, ignored, maltreated; (b) My favorite ideas are attacked by the authors, making me feel insecure, threatened and the infidels must be annihilated.

Sometimes papers have clear methodical errors or missing controls. Here, clear criticism presented constructively, and even rejection, is a gift to the authors.

What to do when you receive a needlessly aggressive report? Read it through; absorb the emotions. Find a secluded place and scream, break old plates if necessary. Return to the report. Notice how hard it is to comprehend the meaning of a comment when language is aggressive. Try to separate the content from the aggressive form. Use this to improve the MS. Use the pain to help you empathize better the next time it's your turn to be a referee.

PS On non-violent communication

Non violent communication, developed by Marshall Rosenberg, is a way for communicating in an affective and authentic way. It can be viewed a vision of the ideal report- were it only so. In a nutshell: Violent language has the template: You are blank, and you should do blank.

Non-Violent language: When I see blank, I feel blank. I need blank and I would like you to blank.

[Marshall Rosenberg, Non violent Communication, Puddledancer Press; 2nd edition (2003)]: