The Philosophers' Cocoon

There are many ways to write a paper. Philosophy would be boring if that weren’t the case. 

Still, I want to register a worry about what I think is an unproductive trend in how people write papers, that is, the increased tendency to “referee-proof” them, addressing all sorts of worries and concerns potential referees might have. I see this more and more in papers I’m refereeing. I used to assume those papers had been through review before, but I think authors are doing it pre-emptively now. I also see referee-proofing offered as explicit advice on how to get one’s stuff published. For example, in a recent article on this blog on how to get published Marcus Arvan writes: 

Finally, I worked on referee-proofing articles, as I found through a lot of trial and error that one really does need to try to foresee and head off “every possible objection” referees can raise to a piece–as, in my experience, if an objection is reasonably foreseeable, some referee will raise it, and it might just be the objection that (if you don’t address it) leaves them unconvinced enough to recommend publication.

It’s obviously a good idea to write in such a way that readers do not misunderstand you and that you make your claim as clearly as possible. There’s nothing wrong with referee-proofing in that weaker sense. 

But I question whether it’s advisable or even possible to referee-proof in the stronger sense of heading off every possible objection, as Marcus supposes, so I will here gently and in the spirit of constructive dialogue (I think!) register my disagreement with him. Here are my reasons why I think referee-proofing is not advisable.

1. As you referee-proof, you sacrifice space to flesh out your position 

One wants to get published and maybe the chance increases due to referee-proofing. Still, referee-proofing is not without cost and one cost is the space one has to devote to head off objections of imagined referees rather than carefully staking out the position.

As an example, here are two papers I wrote (both written while I was not on the tenure track). They aren’t refereed proofed at all yet got into decent journals, this one in Australasian Journal of Philosophy on origins of philosophical intuitions and this one in Philosophers’ Imprint on mathematical realism and the evolutionary origins of maths. I spend practically no time in responding to potential objections, because I want to flesh out the positions I defend there in detail. The time you spend replying to imagined and projected (not actual) objections is time you do not have to develop your position in detail, which readers can then assess and respond to.

Obviously, both papers do take into account certain objections that actual people gave (actual referees, the editors of both journals, friends who read earlier drafts, audiences at conferences etc etc) but I did not try to imagine all possible reasonable objections for myself. This was particularly important for the AJP paper, as the journal has a fairly strict word count policy (at least upon initial submission). PI has more freedom in this respect, but one still tries to keep the word count manageable. I really wanted to flesh out the positions in as much detail as reasonable word count allowed. The platonism paper required numerous references to empirical literature to do so, as well as detailed engagement to Justin Clarke-Doane’s evolutionary anti-realism about mathematics. I could not do that and also respond to imagined referee objections while keeping these papers manageably short. 

2. Can we anticipate objections of potential referees?

The fact that two referee reports will often have very different objections/concerns indicates that it is hard to anticipate which objections referees might have. Alone with one’s own thoughts, one cannot obtain the required epistemic friction that’s caused by interactions with actual people who are responding to one’s ideas (see Medina on getting out of one’s comfort zone by interaction with actual people and situations).

I’m not saying it cannot be done. Thomas Aquinas’ Summa theologia (random sample here) is a nice example of careful addressing of potential concerns. It is also very long. In sum, we risk either picking out some objections that our audience doesn’t necessarily care about or  end up with a very long, exhaustive list of objections. 

3. It adds unnecessary ballast 

I don’t want to be nostalgic about the golden days of analytic philosophy, but whenever I pick up an old paper, say by Donald Davidson on W.V.O. Quine I’m often struck by how terse those papers are. I do think that a more extensive engagement with the literature and inclusive citations are a virtue. But citations don’t add much to the cognitive load/ballast of a paper, as you don’t need to look at them in depth as you read. By contrast, referee-proofing does demand attention from the reader, and I’m skeptical the author is always in the best position to address all the plausible objections readers might have (see point 2). And so, the paper becomes longer but not, I think, always more interesting.

4. It’s a buzzkill

One of the fun things about reading a philosophy paper is the epistemic emotions that the reader experiences as she works her way through the paper. Hey wait a minute, is he really saying that? What about the Euthyphro dilemma? And so on. A really obvious blatant potential objection should of course be addressed to assure the reader that the author has considered the objection and is dealing with it. But to see — as referee-proofing tends to produce — a whole bunch of caveats and replies etc for every claim kills one’s buzz as a participant in the reading process. It reduces the reader from an active audience to someone who needs to be told what the main objections are. It makes reading philosophy less participatory. 

5. Lots of philosophy is not concerned with objections, why should each paper address them exhaustively? 

I think that as a discipline we should aim to expand the ways in which we do philosophy, not narrow them down. To make this point fully I need another blogpost, for another day, but in brief, I think stories can be philosophy, aphorisms can be philosophy, pictures can be philosophy, poems, dialogues etc. And this is not as bizarre as it seems, as a lot of philosophical work that fits in one of those formats, e.g., Confucius, Zhuangzi, Ibn Tufayl, Hume, Wittgenstein. This is playful, narrative, expansive philosophy.

Yet in the quest to get papers into journal space (and of course, grad students, postdocs, etc as well as more established faculty will want to publish) there is a homogenization in how philosophical papers is written even in the last decades. You have the 150 word abstract, the brief intro ending with a bird’s eye overview of what the author will accomplish in sections 2, 3, 4 etc. Referee proofing, if it becomes the norm, becomes yet another way in which philosophy papers formally resemble each other. I can see the rationale for this standardization but we do sacrifice such goods like surprise, wonder, in the process. 

In sum, addressing obvious, blatant objections is fine, and addressing objections by actual referees is necessary, but beyond that, I do not think bringing up all sorts of objections for the sake of bringing them down is beneficial in philosophical writing.