In the past few weeks I've had several peer review requests and it has always struck me that it is far too easy to come across as a heartless bastard when blind reviewing work. That's why I have decided to always be kind in my reviews, even when offering the harshest of criticism.
Consider the following humorous text that has been doing the rounds on Twitter:
Dear Sir, Madame, or Other:
Enclosed is our latest version of MS #85-02-22-RRRRR, that is, the re-re-re-revised version of our paper. Choke on it. We have again rewritten the entire manuscript from start to finish. We even changed the goddamned running head! Hopefully we have suffered enough by now to satisfy even your bloodthirsty reviewers.
I shall skip the usual point-by-point description of every single change we made in response to the critiques. After all, it is fairly clear that your reviewers are less interested in details of scientific procedure than in working out their personality problems and sexual frustrations by seeking some sort of demented glee in the sadistic and arbitrary exercise of tyrannical power over hapless authors like ourselves who happen to fall into their clutches. We do understand that, in view of the misanthropic psychopaths you have on your editorial board, you need to keep sending them papers, for if they weren't reviewing manuscripts they'd probably be out mugging old ladies or clubbing baby seals to death. Still, from this batch of reviewers, C was clearly the most hostile, and we request that you not ask her or him to review this revision. Indeed, we have mailed letter bombs to four or five people we suspected of being reviewer C, so if you send the manuscript back to them the review process could be unduly delayed.
Some of the reviewers comments we couldn't do anything about. For example, if (as reviewer C suggested), several of my ancestry were indeed drawn from other species, it is too late to change that. Other suggestions were implemented, however, and the paper has improved and benefited. Thus, you suggested that we shorten the manuscript by 5 pages, and we were able to do this very effectively by altering the margins and printing the paper in a different font with a smaller typeface. We agree with you that the paper is much better this way.
One perplexing problem was dealing with suggestions #13-28 by reviewer B. As you may recall (that is, if you even bother reading the reviews before doing your decision letter), that reviewer listed 16 works the he/she felt we should cite in this paper. These were on a variety of different topics, none of which had any relevance to our work that we could see. Indeed, one was an essay on the Spanish-American War from a high school literary magazine. the only common thread was that all 16 were by the same author, presumably someone reviewer B greatly admires and feels should be more widely cited. To handle this, we have modified the introduction and added, after the review of relevant literature, a subsection entitled "Review of Irrelevant Literature" that discusses these articles and also duly addresses some of the more asinine suggestions by other reviewers.
We hope that you will be pleased with this revision and finally recognize how urgently deserving of publication this work is. If not, then you are an unscrupulous, depraved monster with no shred of human decency. You ought to be in a cage. May whatever heritage you come from be the butt of the next round of ethnic jokes. If you do accept it, however, we wish to thank you for your patience and wisdom throughout this process and to express our appreciation of you scholarly insights. To repay you, we would be happy to review some manuscripts for you; please send us the next manuscript that any of these reviewers sends to your journal.
Assuming you accept this paper, we would also like to add a footnote acknowledging your help with this manuscript and to point out that we liked this paper much better the way we originally wrote it but you held the editorial shotgun to our heads and forced us to chop, reshuffle, restate, hedge, expand, shorten, and in general convert a meaty paper into stir-fried vegetables. We couldn't or wouldn't, have done it without your input.
[Name Removed for Blind Review]
Now, it is, in my opinion, very easy to fall into to the trap of becoming a monster behind the shield of anonymity. I would wager that even the most terrible 30 page paper that I've been asked to review had taken an enormous amount of work on behalf of the author and, even though I couldn't in good conscience recommend it for publication, I also couldn't give a response that abused my position in order to abuse the author.
What I'm saying is this: do not be a soft touch, but be courteous and empathic. I'm still a tough reviewer and I won't tolerate people getting it wrong or being loose in terminology. However, I'm not a pedantic reviewer. I don't insist on my own research being there in every paper and I don't insist on everything being worked out to the 100% end of the road; that's the point of the scholarly process. I am also not a rude reviewer; I give feedback that is designed to improve the work with suggestions for future directions, even if that is pointed out through failings in the current mode. Do not allow yourself to forget that there is a human on the other end, because of anonymous, technological mediation.