I'm dancing. I'm so happy, if you remember, because my paper was rejected – but I know it's a good rejection that I can change, as if by magic (but really by hard work and some luck).

Rejections are part of what we do, but they are certainly not unique to our profession. There's a story that the actor Brad Pitt experienced eighty-four consecutive rejections at the start of his career before finally landing his first job in a commercial for a fast-food chain. And that's Brad Pitt. It doesn't make rejection of our grants and papers any easier of course. Rejection hurts.

Last time, we talked about how to read a rejection, how rejection isn't always final. So now you're faced with writing the rebuttal, which you absolutely, positively must not do on the day you were rejected.

By taking some additional time to review the criticisms, you may have realized that some of them may even be valid. Even if they aren't, act like they are. A major problem with peer review is the peer part. Most of the people who are really competent to evaluate your work said no, they were too busy. Even if they weren't, they said no because they just didn't want to commit to the job. Or they said yes, and then handed the assignment to a graduate student, who thought that this was a homework assignment to find anything they could imagine to make the author's life miserable. (If you, as a reviewer, pass off these reviewing jobs on others, shame on you. It just shouldn't be done. You were asked to review and you accepted the responsibility – even if you want to educate your students, you can have them try their hand after you've done it. If you feel that your trainees are fully qualified to do the review, then tell the editor or grant administrator that you would like that person to be have the job. If they reassign, fine, but don't lead them to believe that a review written by a trainee was done by you. It is simply not appropriate.)

Right, where was I? Oh yes, the critiques. It is generally useful to write out, in advance, how you plan to address each point. It's painful, but it's even more painful after a few weeks of work to realize that you left out the control one of the reviewers wanted you to do. Plan carefully, because the point here is to address the concerns head on. And if you can do this with more experiments, this is very much to your advantage. This is for a few reasons, but the main one is this: it means that the observations you have made and are reporting are reproducible. This lends confidence to the process, and it will score you points. It is rare that a rejection can be turned around without more experiments – you won't be able to simply argue that the criticisms are all stupid, for reasons we'll come to.

So now you've done more work, and you reckon that you're going to be able to answer all of the criticisims. The paper or grant has been rewritten and. hey, it's actually better. So now you're going to craft a letter on which everything rides. Don't underestimate the importance of this response: it's the key to turning the rejection into backslaps, parades, and happy toasts (in time). Often, it will be the only thing the reviewers read.

To write the rebuttal, you have to look at this from the perspectives of the reviewers, who donated their time to critique your work. Thank them (even if you really do hate them). They are the ones you need to get on your side. This is not going to happen if you insult their competence or point out their mistakes. Of course they were mistaken, because they rejected your work (not you), and, however much you want to call them idiots, this is not a way to win their hearts and minds.

So here's what you do. First, copy their critiques verbatim (generally easy now, since they usually come electronically, as if by a robot hand lost in the ether). Then after each heading (as in `Reviewer 1') grit your teeth and write, in another color or typeface, this: `We thank the reviewer for his/her careful evaluation of our work. In addressing these valid concerns we feel we have substantially improved it.' Yes, you're groveling. You're going to do a lot of that. Do you want to turn your rejection into an acceptance, or not?

Then, after each and every criticism, detail exactly what you've done to address it, again in a different color or typeface. This is so they won't miss it. Even if you think you didn't have to do anything because they misunderstood you, the fact that they misunderstood you is your fault, and you have to say so. `The reviewer notes that we didn't include any data in our paper. We apologize for representing our data as figures that were appended to the end of the manuscript. We have now drawn the reader's attention to these in the text.' Even if you don't change anything, pretend that you have. They'll realize their mistake, but you cannot rub their noses in it.

Often it will happen that the comments of the second, third and nth reviewer simply reiterate the comments of the first, but this doesn't matter; address them anyway. Don't refer Reviewer 2 back to your response to Reviewer 1, just state your response again – as many times as you need to. Reviewers only rarely read the other reviews or your responses to them. And you don't want to send them scurrying around to other pages in your rebuttal when they are reading the one addressed to them.

Whenever possible, thank them for the criticism, as indeed you should (sometimes). The only exception to this might be in the correction of a typographic error, but even here it doesn't hurt. When they've picked up a mistake you've made, thank them for catching it. When they've suggested an experiment that actually turned out to be informative, tell them so. Get them on your side. They'll have to be if you're going to make this happen.

In some cases, it won't be possible to address a concern in a manner that actually addresses the concern. You might have tried the experiment, and it didn't work (but you want to argue that it doesn't matter). Or the experiment is impossible. Or very, very difficult – too difficult for this submission. What to do? Here, groveling is especially important. `We thank the reviewer for his/her suggestion that the key experiment be repeated at zero gravity. While we agree that this would certainly be of interest, we respectfully suggest that this might be suitable for a future study based on our observations, and have not addressed this point in the resubmission. We will certainly endeavor to perform this experiment the next time we are in free fall.'

Finally, as awful as it sounds, you have to thank the reviewer again, because you have to beg them to change their minds. `Again, we thank the reviewer for the effort that has gone into evaluating our work, and hope that in addressing each of these points, he/she now agrees with us that the submission is substantially improved.'

Yes, you feel like the worst sort of kisser-upper. But again, put yourself in the place of the reviewer. He or she is busy, and because they took time out of their busy day, some time ago, to read and comment on your paper or grant, they are now being tapped to drop everything at their earliest convenience and read your revision. It will not have come at a welcome time, because it never does. In fact, because you were anxious to get this done before you went on holiday, it is falling into their laps on their holiday. They are not amused by this. So if you aren't hugely thankful, you should be – or at least, that's how they feel. And maybe, by being very, very nice to them, you might cheer them up a little, because they can just check over what you've written in your rebuttal letter without having to go back and redo the whole thing.

But even if you've tried your best, even if you've addressed all the issues, it can still happen that you will once again crash and burn, and maybe this time it will be a just go away rejection. This is so terribly painful that I am hurting for you already. Although it won't help at all, you might consider what my wonderful friend Professor Hedgehog once did. His paper had been rejected by a rather desirable journal, and rather than responding angrily, he wrote them a letter that went like this:

Dear Editors,

Thank you for the rejection of our paper. As you know we receive a great many rejections, and unfortunately it is not possible for us to accept all of them. Your rejection was carefully reviewed by three experts in our laboratory, and based on their opinions, we find that it is not possible for us to accept your rejection. By this we do not imply any lack of esteem for you or your journal, and we hope that you will not hesitate to reject our papers in the future.

Yours sincerely,

Professor Hedgehog.

Rejection ruins our days, makes us bitter, squashes our hopes and happiness, and takes the joy out of the joyful things we do. My friend's letter wasn't as good as getting my paper published or my grant funded, but it made me dance anyway. Even if you get your paper accepted eventually, even when your grant is approved, the next ones will go down in flames. So, let's dance anyway. It's how we live. Remember Alan Bates at the end of Zorba the Greek? His project, the thing that has driven him and the villagers for weeks, falls to ruin around him. Desolation and despair. And he rises, and turns to Anthony Quinn, and shouts, “Zorba, teach me to dance!”