Saturday, February 21, 2015

How I Write for Peer Review

Publish or perish, or so they say. That’s the rule in academia. But not all publications are created equal. I’ve “published” over 700 posts on this blog (and republished many on other blogs), and although I think there are advantages to having done so, I’d be lying if I said these publications were academically “significant”. They’re certainly not significant from the perspective of the administrators and overseers lurking within the groves of academe. If you want to please these people you must produce peer-reviewed publications (preferably double or triple-blind peer-reviewed publications) in high impact academic journals. That’s where the game is.

This is not to disparage the system. Suffice to say that, despite my rather cynical and world-weary tone in the opening paragraph, I’m something of a fan of peer review. With some limited exceptions, I can safely say that most of the articles I’ve published in academic journals have been improved by the process. But, in any event, my focus here is not on the merits or demerits of peer review. Instead, my focus is on the practical matter of how one actually goes about producing and publishing material in peer-reviewed journals.

Or, more precisely, my focus is on how I go about doing this. I focus on myself not because I am a self-obsessed egomaniac (though I may be) but because I don’t think there is a definitive set of “how tos” for success in peer review. Different people will have different methods, some of which may work for them but not for others. What I offer here is just my approach. I offer it in the hope that it will be useful to some, and in the spirit of transparency.

I’m not particularly successful in producing peer-reviewed publications (probably 12 good pieces in the past 4 and half years, and 4 other less good pieces), but I have some experience that might be worth sharing. One thing I noticed as a struggling PhD student — and this may have just all been in my head — was how guarded and competitive people seemed to be about their attempts at publication. It’s as if everyone knows that everyone else is grinding away at it, but no one is willing to share the stories of their failures. Given the levels of neurosis and impostor syndrome in the academy, it would be nice (I think) if people were more open about these things.

So in that spirit of openness, here are the steps I generally follow when writing for peer-review.

Step One: Prepare for Failure
Trying to publish in academic journals is often a soul-destroying process. Acceptance rates at the top journals are absurdly low (under 5% in some cases), and even at middle-ranked journals one is far more likely to be rejected than not. I might like to think that I am special, that my paper is truly exceptional and that the peer reviewers will be overwhelmed by the subtlety and sophistication of its argument, but the likelihood is that I am wrong. I have now come to accept this, and before I even start writing I prepare for failure.

I put this as step one because I think coping with failure is the most important thing to master. It’s certainly something I wish I had been better able to deal with before I embarked on an academic career. Throughout my school and undergraduate days, I was used to success. I had never failed an exam; I was (nearly) always at the top end of my class. The notion that something I had written would be rejected — often with peer-reviewers questioning my basic competence and understanding — was completely alien to me. Now, I’ve learned to deal with the fact that it will be a fairly regular occurrence.

Part of the trick is, I think, to frame failure in a positive way. In his book How to Write a Lot, Paul J Silva suggests that your goal should be to become the most rejected author in your department. Why? Because being the most rejected means you are also likely to be the most published. The reason for this is twofold. First, if you are throwing a lot of darts at the dartboard one of them will eventually hit the target. Second, most papers that get rejected from one journal will end up being published elsewhere. Although it may be painful and disappointing at first, taking onboard the reviewers’ criticisms, revising the paper and submitting it elsewhere will often result in success. That’s true of the most recent paper that I had published. Indeed, one of my biggest regrets is that I lacked persistence with some of my earlier papers. In 2010 I had three papers rejected from journals that, in retrospect, could have been published elsewhere. But I just gave up on them after one round of rejection.

Step Two: Generating Ideas
Although writing for peer review can seem like a tedious and grubbily instrumentalist process, it’s important to realise that at its foundation there is (one hopes) some genuine intellectual passion and interest. That’s certainly true for me. I write papers about topics that interest me (even if only for while) and generating ideas for papers is something that provides me with great joy.

Unfortunately, I have no idea how I manage to generate such ideas. I don’t think I have ever generated an idea for a paper through brute intellectual force (i.e. by simply sitting down and demanding myself to come up with an idea). They just seem to occur to me. That said, there are some discernible patterns. I usually come up with ideas when out running or cycling (when I’m free from other distractions), or else late at night (typically between 11:00 pm and 1:00 am). I don’t know why this is. Most of my ideas come from combining and manipulating arguments that I have found in other papers that I have read. I don't think I've ever had a truly original thought. For example, my paper on sex work and technological unemployment came from combining ideas that I came across when reading on the ethics and economics of prostitution with other ideas I encountered when reading about robots and technological unemployment.

By an “idea” for a paper I mean a basic argument I wish to defend. Sometimes those ideas come to me in a conclusion-first format. That is to say, I’ll think “wouldn’t it be interesting if someone were to argue X” and then work backwards from there. Other times, the ideas come in a premise-first format. That is to say, I’ll think “what does principle X imply?” and try to work out the details. Again, I don’t know how or why things occur to me in these formats. I believe that wide reading and ongoing curiosity are the key, but I don’t have a precise algorithm or strategy.

Step Three: Writing the Abstract
Once I think I have the kernel of a good idea (and this idea could be percolating around my brain for several months before I take it any further), I will try to write the abstract for the final paper. Some people might find this odd, but it is a habit I picked up from reading Wendy Belcher’s book Writing Your Journal Article in 12 Week, which I recommended on a previous occasion. As she notes in that book, a good abstract should provide a summary of the argument you wish to defend in the paper. To me it makes sense to try to generate that summary first. Hence why I write the abstract first. I may, of course, revise it at a later point, if my argument changes slightly.

To help me to write the abstract, I will often use Stephen Posusta’s “Instant Thesis Generator” (something I also picked up from Belcher’s book). This comes from something called the Procrastinor’s Guide to Writing an Effective Term Paper, which as you may gather is a book for lazy undergraduates. Again, it might seem odd that I would rely on this, but I find it suprisingly beneficial because it provides a useful heuristic for thinking about a good argument. It is illustrated in the diagram below.

You can click on the image for a fuller explanation, but the basic gist of it is that you should be able to summarise your paper using a fill-in-the-blanks sentence: “Although….Nevertheless…Because”. The first part is a statement of the contrary point(s) of view (the ones you will be opposing in your article); the second part is a statement of your main conclusion(s); and the third part is a list of the reasons, evidence and argumentation you will use to support your conclusion(s). Of course, I will rarely actually write the abstract out in this exact format. But I will use it to structure how I think about the abstract. In essence, I think that if I can’t summarise my paper using the instant thesis generator, I probably don’t have an idea worth pursuing in any more detail. It may need some further percolation.

Step Four: Planning the Article
Once I have the abstract in place, I will proceed to write out a more detailed plan for the article. I will do this using pen and paper, and I will usually do it late at night (after 11:00 p.m.). For whatever reason, I find it easiest to work on plans at this time. I believe this is because article planning is, for me, an exercise in sustained thinking, and I find it easier to do this when I am free of all other distractions (emails, work, cooking etc.).

Unlike generating article ideas, I find that planning an article is something that can be done through brute intellectual force. Most articles I have written have been planned out in about one to two hours, by literally just sitting down and forcing myself to think through each step of the argument I wish to make. In other words, using my abstract as a starting point, I force myself to think about each of the premises I need to defend, the counterarguments that are likely to be thrown at me, and my responses to them.

My plans are pretty detailed. For a 10,000 word article, I’ll usually scribble out anywhere between 4 and 10 A4 pages-worth of a plan. I do this mainly because I like the actual writing process to be as mechanical and methodical as possible. Simply a question of filling in the details and smoothing out the edges of the plan.

Step Five: Writing the First Draft
With a detailed plan in place, I can proceed to sit down and write the first draft. In many ways, I find this to be the easiest and possibly least exciting part of the process (though there is a nice sense of accomplishment that comes with it). I follow pretty much the same timetable for writing all my articles. I set aside about two hours every morning (usually between 9 and 11 or 10 and 12), and I write approximately 1,000 to 2,000 words every day. In this way, I typically manage to produce a first draft of an article in about two weeks. Obviously, it all depends somewhat on the time of year and the presence of other distractions, but I think I can genuinely say that every one of my published articles had their first drafts produced in a two-week period.

Step Six: Revising and Seeking Feedback
Once I have finished the first draft of the paper, I will set it to one side for a period of time. The period of time varies. Sometimes it will be a month; sometimes it could be six months. Ideally, I aim for the shorter end of this spectrum: if it takes me six months to get back to a paper, that’s usually a sign that I didn’t really like what I wrote and I need more distance from it. Anyway, once I return to the paper, I will read through it and revise it however I see fit. This process usually takes only a couple of hours, though it may take longer if I think the paper needs substantive revisions. When revising, I do the usual things: I look for weakpoints in the argument and think of possible amendments (oftentimes I’ll have flagged these in bold in the initial draft), I try to improve the overall “flow” of the writing, and I try to correct for any spelling mistakes and referencing errors/omissions. I am, however, notoriously bad at that and nearly everything I have ever published (including articles that ultimately pass through copy-editing and proof-reading) will have some typographical and spelling errors.

At this point I face a choice. I can either send the paper around to some colleagues for feedback, or I can submit it to a journal. I’m guessing the former is preferable, and it seems to be a common practice among some of my peers, but I don’t do it that often. I usually send things out to journals before seeking feedback from others. Indeed, even though I have participated in various “work in progress” seminars over the years, in most cases the papers I was discussing had already been submitted.

I think the reason I prefer to submit without soliciting feedback is that I’m impatient and I’m all too aware of how slow the peer review process can be. I’m also a little bit insecure. I’m now pretty good at taking critical feedback from anonymous reviewers but I'm still less good at taking it from people I know. Once I’m happy enough with a piece I’ve written myself, I just want to get the peer-review ball rolling.

That said, once I’ve submitted the piece I will sometimes seek feedback (in anticipation of the need to revise once the journal gets back to me), or else I will seek feedback once I have received a “revisions requested” or “revise and resubmit” notification from a journal. This may, once again, reveal my insecurities: I’m less bothered by what people I know say about what I’ve written if I have some affirmation from the peer reviewers.

Step Seven: Submission
I’ve jumped the gun to this step already but just to reiterate, once I’ve revised the paper and I’m happy with it myself, I will submit it to a journal. I’m not hugely strategic about the choice of journal. I want something with a decent reputation and an international audience, but beyond that I’m not too fussy. When I’m writing I will usually have a number of options in mind (in fact, if I’m honest, I will rarely proceed with writing a paper if I can’t think of possible target journal in advance), and I’ll start by picking the best of those and then working my way down the hierarchy afterwards if needs be.

Once I have submitted, I’m forced to play the waiting game. Unfortunately, this waiting game can take a long time. In terms of my own publications, the best experiences I have had when it comes to the time from submission to initial decision were with Neuroethics (accepted with revisions), Religious Studies (accepted) and Journal of Medical Ethics (rejected) all of whom got back to me within a month and half. The worst experiences I’ve had, alas, were with Sophia and Law and Philosophy, where in both cases I was waiting nearly nine months for an initial decision. That said, both pieces were ultimately accepted and I know that these are far from being the worst waiting times. There may have been good reasons for them too (e.g. waiting for reviewers to write up their reports).

Step Eight: Prepare for Failure Again
When I have submitted the paper I will prepare myself, once again, for the possibility of failure. I think it is important to remind myself of this after the initial submission because, once I have written the paper, I may be lulled into a false sense of immodesty. I’ll think that what I’ve written is wonderful and important and bound to be accepted. It’s important that I disabuse myself of such notions. If I don’t, the reviewers of my paper will be quick to do so.

Step Nine: Responding to the Decision
…eventually I’ll receive a decision about my paper. If all goes well, the paper will either be accepted outright (this has never happened to me), accepted with revisions (major or minor) or I’ll be asked to revise and resubmit. If I receive one of those latter two decisions, I will parse the reviewers’ comments immediately and then I’ll go through the five stages of grief: denial (“they couldn’t have said that about my paper”); anger (“the reviewers must be idiots”); bargaining (“if only I could get them to see this point”); depression (“that’s it, I’m never writing another paper again”) and, finally, acceptance (“well, I suppose I better get on with it and revise the damn thing”). In the early days, it would take me a while to move through those five stages. Nowadays, I’m usually onto stage five within 24 hours.

I find responding to the reviewers to be one of the more difficult parts of the process. Unlike the writing of the first draft, I can only manage revisions in concentrated bursts. I usually manage them over two to three days (normally at weekends during teaching semesters). I’ll sit down on day one, read through my paper and the reviewers’ comments and try to come up with some strategy for responding to them. Then, on day two, I’ll add my revisions to the paper, taking in a third day if necessary. By way of illustration, on my two most recent papers (“Robotic Rape and Robotic Child Sexual Abuse” and “The Normativity of Linguistic Originalism”), I managed to complete the revisions over two and three days, respectively. Ideally, I like to do this pretty soon after I receive the comments, but sometimes it takes me a while to schedule the two or three days I need. I have no tips to offer on making your revision more likely to be accepted. The only thing I do — which I suspect everyone does — is to prepare a separate document with a very detailed response to each of the reviewers’ comments. The goal of this is to show how I have taken onboard everything they have said, and where exactly in the paper I address the point. In this document, I will say things like “on the penultimate paragraph of pg. 22, I have addressed reviewer one’s point about X by saying the following”. This level of precision and detail is something I picked up, in particular, from Matthew Kramer, who was a reviewer on a paper I once wrote.

If the paper is rejected (either initially or after revisions), I will also go through the five stages of grief, with the main difference being that the acceptance stage either consists of: (a) deciding to revise the paper and submit it elsewhere; or (b) abandoning it completely. As I mentioned at the outset, I’m less inclined to go for the nuclear option of (b) now, though it used to be my norm. That said, I am tempted by it on occasion. For example, I’m currently in the process of submitting a paper I wrote to a third journal, after it was rejected twice before. If it is rejected a third time I will certainly abandon it completely. The world is must be trying to tell me that it’s not a very good idea.

So that’s it. Those are the nine steps I usually follow when writing for peer reviewed journals. I’ll close with my favourite comment from an anonymous reviewer of one of my papers (if you’re interested it was about this paper). The comment is a positive one, but in a backhanded way that is typical among certain philosophers:

“This is a good paper. In the opinion of this reviewer, it is wrong at nearly every important point, but it is wrong in ways that are interesting and important -- a genuine contribution to the philosophical discussion.”

It almost makes it all worthwhile.

No comments:

Post a Comment