How do you write?

I thought I might follow up my post on reading philosophy with a post on writing! Here again, I'll share my approach, and then ask you all to share yours. It'll be interesting to see, I think, how different people go about the writing process. Okay then, here goes.

My writing process has changed immensely over the years. As an undergraduate, and for most of my career as a graduate student, I absolutely slaved over papers, composing them slowly and carefully, crafting and revising every sentence as I went along. When I was writing (typically, term-paper time), I would literally spend all day, just about every day of the week, for weeks or even months on end, to get the paper just right. Unfortunately, at some point in grad school -- unsurprisingly, I think it was around the dissertation time -- this process stopped working effectively. My production ground to a halt. I think, like many grad students, the dissertation seemed like such a monumental, herculean task that carefully crafting one page at a time rapidly transformed into crafting no pages at a time. ;) 

Since that time, I've had three particularly formative experiences, each of which dramatically changed my writing process, and which I'd like to share now.

I've shared the first of these formative experiences several times before. As luck would have it, after a couple of years of getting nothing done on my dissertation, I received an unsolicited self-help book in my department mailbox on how to write a dissertation. Although I normally have no interest in self-help books, I was so desperate with my dissertation at that point that I said, "What the heck!", and actually read it. Its main piece of advice was this: make yourself free-write some small number X (e.g. 3-5) pages per day -- never any more, never any less -- with absolutely no self-censorship or editing, and preferably first thing in the morning. The advice then allowed you to do other stuff (editing, revising, etc.), but only after you'd completed your mandatory free-write.  The basic rationale behind the advice is that the most crucial things with writing are to (1) actually force stuff out of your head rather than procrastinate, and (2) maintain a positive daily attitude where you actually wake up each day wanting to write. 

My experience following this advice was astounding. I can still remember the very first week I did it. For about 18 months prior, I'd gotten basically nothing complete with the dissertation. I was always doubting everything, scrapping stuff I'd written as garbage. Then, that first week trying the book's advice, I did it. I got 5 pages of garbage out of my head in a couple of hours the first day, then the same thing the next day, the same thing the third day, etc. By Friday I had 25 pages of garbage. Then, on Saturday, I took that 25 pages of garbage and whittled it down to about 10 pages of really good stuff. I was sold. 9 months later -- after following the advice to a "T" -- I finished and defended my dissertation. Like many dissertations, it wasn't great. But whatever. I got it done. I later gave the book to a friend a year behind me in the program who was also struggling immensely (she basically had none of her dissertation written either). She followed the advice and, just as I did, finished her dissertation about 9 months later. Crazy. I suggest that anyone who finds themselves struggling try the same.

My second formative experience occurred during my first year out of graduate school, during the year I was a VAP at the University of British Columbia. Although the writing strategy above was still working for me, I wasn't publishing. So, I picked up the phone and called a couple of friends from grad school had been publishing up a storm (and in great journals). I asked them what their secret was. They both said the same thing: they write and send out a ton of work, with around 10 articles under review at journals at any given time. Basically, their rationale was this: acceptance rates are so low, and reviewers so unreliable, that this is the only rational way to "play the odds." In their view, even if a particular article is great, chances are it will get rejected several times, in which case if you only have a few papers out for review at any given time, you may not publish anything for years -- again, even if your work is awesome (side note: it's public knowledge that some very famous articles in philosophy were rejected by numerous top-ranked journals before being accepted anywhere). So, anyway, I followed their advice. Instead of wasting a bunch of time "polishing" papers I already had, I basically drafted paper after paper, revised them a bit, and sent them out. Like them, I began publishing up a storm. And, far from leading me to do work of lesser quality, I believe the process of continually drafting and sending out papers led to an increase in the quality of my work. Of all the papers I have published to date, I do not think any of them would have been substantially better had I spent substantially more time on them. Good ideas are good ideas, bad ideas are bad ideas. The quicker you get them out of your head and out to conferences and journals, the quicker you find out which are which. Over the past 6 years, I've drafted well over 30 papers. Only 10 of them have been published. Not a great batting average to be sure (though I think a few of the unpublished ones are very good), but in my experience this is pretty much par for the course. Many people who engage in creative endeavors -- authors, artists, etc. -- will tell you that the vast majority of their work is garbage. As Linus Pauling is famous for saying, "The best way to have a good idea is to have a lot of ideas." Indeed, if anything has been my experience, this has.

My third formative experience was the day, during my first year at Tampa, that my hard drive died and I lost 6 months worth of work (note to you: back up your work!;). I had just spent the previous six months writing, and then polishing, two papers -- both of which I went on to later publish. Indeed, I had literally just put the finishing touches on both papers the day my hard drive died. I was devastated. I couldn't believe that I had lost 6 months of work. But I wasn't about to accept that. So, I went to Best Buy, bought a new laptop, and resolved to rewrite both papers from scratch as quickly as possible, before I forgot everything that I had written. I rewrote both papers from scratch in two days or so, revised them quickly, sent them off, and they both got accepted. And somehow, from that point on, I became able to write papers insanely quickly. As "unlucky" as losing my hard-drive was, in retrospect it now seems like one of the luckiest things that ever happened to me.  These days, I find myself writing first drafts in a few days to maybe a week or two tops. Some days, I'll draft 5 pages or so. Other days I may draft anywhere between 15-20 pages. Revising, obviously, takes much, much longer. But still, I typically get full drafts done super quick now. And apparently I'm not alone. I've heard similar things from some other people I know, as well as from my wife, who works in a different field (her mentors say that they, too, tend to write paper drafts in a couple of days). Just a few years ago, I would have never imagined writing papers so quickly -- and I certainly never would have imagined it in grad school. But that's the way it is now. Somehow, over time, one just becomes much more efficient -- and again, I don't think it's come at a cost of quality.

Finally, I guess I will share in a bit more detail what my actual writing process looks like. When drafting, I tend not to have any notes or books around. I also do not do outlines (I never have). I more or less just sit down and start writing off the top of my head. If I don't know the literature well enough to do that, as far as I'm concerned I'm not ready to write. So, I only sit down to write once I feel like I know the literature well enough in my head to situate my work in it -- and, because of this, I don't worry about references (and such) at the drafting stage. I just write. I get the introduction out quickly, then summarize whatever I need to in the literature, and then just lay out my argument, consider objections, and go from there. When I told my wife this is how I draft -- without books, without notes -- she was very surprised. But then she asked her mentors (again, in a very different field) how they draft, and they said they do it the same way. And she works in a field that has very strict citation requirements. So, it seems to me, what I do is apparently what other people do too. Getting bogged down with books, articles, and notes just slows me down too much at the drafting stage. Again, as far as I'm concerned, if I can't lay out my project and situate it in the literature off the top of my head, I'm not ready to begin writing. It's only after drafting that I will go back, sentence by sentence, and section by section, with books, articles, and notes to add in citations, deepen my discussion, etc. Anyway, that's my drafting process: quick and straightforward. Phenomenologically, it feels a lot like laying brick work or planks in putting a boat together. I just lay things out as simply and straightforwardly as I can, and fill in all the details later.

My revising process is much more labor intensive. After I have a complete draft (and only then), I will go back and start reading my work at page 1. Here, I try to make each sentence as clear and lean as I can, cutting away unnecessary words and chopping up overly long sentences into multiple shorter ones. I also put in citations, and try to clear away unnecessary italics (which I have a really bad habit of overdoing when I write;). If, say, I get to editing/revising page 5 on a given day, I will not start at page 5 the next day (there go those italics again!). Rather, the next day, I will start once again at page 1, read all the way up through page 5 (making any necessary revisions along the way), and then engage in the labor intensive revisions from page 5 forward. Let's say, then, that I make it to page 10. Do I begin at page 10 the next day? No! I begin, once again, at page 1. It drives me a bit batty, but each time I do this I discover things I don't like, and end up making changes. In this way -- by starting my revisions at page 1 every single time -- I force myself to repeatedly read, assess, and fix earlier revisions. Which I think is a good thing.

That, more or less, is how I revise. What about feedback? I've heard many people say that getting good feedback on work is crucial. I sort of agree, but for reasons I'll give below, I'm actually of two minds about it. I've always been pretty bad at seeking out feedback. First, believe it or not, I'm rather shy, both intellecually and socially. I always get nervous showing people my work, and so, all too often, I just avoid doing it. Second, I also work in an extremely small department -- there are just 3 of us, and we work in different areas -- so it is hard to seek out feedback locally.

In some cases, I think my difficulties getting feedback has worked against me. It took 6 years for me to publish my paper, "First Steps Toward a Nonideal Theory of Justice" -- and, while I'm thankful that the process turned out the way it did (I think it really did take all that time to become the paper it should have been), I suspect it might not have taken nearly so long had I been better about seeking our feedback (almost all of the feedback I received on it came from journal reviewers, as well as a few conferences). In other cases, however, I think my not seeking out much feedback has worked in my favor. Early in my career, I was easily discouraged, especially when working on really ambitious ideas. While it is important to get, and be open to feedback, I also think it is very important to learn to trust oneself as a philosopher -- and I think working more or less alone has helped me in this. In short, I think "the right amount of feedback" is a fine line. If you don't get enough feedback, you may end up toiling away on bad ideas -- ideas that feedback could set you straight on. On the other hand, too much feedback can be stultifying. My feeling is...you sort of have to feel it out, and see what amount and what kinds of feedback work best for you.

Anyway, this is how I write. How do you?

Comments

[Jon]

I just discovered your blog and really enjoyed reading this post. I look forward to exploring around elsewhere on your blog now. I've always been interested in how other people go about the writing process so this was interesting to see. Particularly since I'm an early career prof. working on a the first paper of a new project post-dissertation. I have been grinding away at it the way you described doing early on...I may just try getting the rest of it down on paper and working from there rather than being so meticulous with each paragraph as I plod along.

Thanks for sharing!

[Anthony Adrian]

Not sure how many people would be interested in how I write since I haven't published anything, but here goes nothing:

Everyday I wake up at or around 6 a.m., make a pot of coffee and plop right in front of my computer to write for about two hours. It's the only time I feel creative. Maybe it's the caffeine. Maybe my mind still has one toe in the pool of the subconscious. Who knows? At minimum, writing first thing in the morning affords me the all too important opportunity to indulge in my coffee addiction while being productive.

But after plopping down, things go one of two ways. The first way—if my idealized, disciplined self has his way—ends up with me writing using the Mumford method. It's divine. Really. You'll be churning out papers in no time by taking his advice. His approach to writing is just about bullet-proof, except where it's not.

My beef with Mumford's method comes down to this: it doesn't allow me to practice my prose! This is important to me because...I'm not so good at writing prose. But I aspire to be better. The thing is, if you are to follow Mumford's advice, you will work on a paper in bullet-note format until the idea is rock solid, write one complete draft, fix spelling errors and turn the thing in.

At what point am I supposed to develop my style? As I write the first draft (which doubles as the final draft)? No. After I've turned the paper in? (In my case, for a grade.) Again, no. It seems as though you're left with one option: fiddle with the style after completing the the first/final draft but before turning it in. But this takes away a good deal of the allure the method once had. Mumford suggests his routine allows one to produce more work with minimal revisions. But producing stylish work is just so damn hard. I feel as though I need to practice it every day, not once a paper is in its near-finished form.

The second, and more frequent, way things turn out involves me sitting down and writing the way Mumford wants me to avoid: I type up a bunch of stuff in my text editor (Emacs, of course), "thinking" as I write. The scare quotes are there because I don't always see it as thinking an issue through for the very first time. Instead, I think of it as practicing my philosophical skills simpliciter. Consider it this way: in nature a substance is never without properties, though we may "abstract" the substance from its properties, and vice versa. Likewise, in nature, content is never without style, though we may abstract the content from the way in which it is presented, and vice versa. Given my pathological aversion to using symbols—while being a member of the Society for Exact Philosophy!—this means there's plenty of style issues for me to work through. Sure, lots of the stuff I write gets the old heave-ho. But I write in LaTeX (sometime Markdown) so the material I nix remains with me, not just in my heart but also in the document, only "commented" out.

One of my professors—himself a style monger—claimed Mumford's method is a great tool for people who don't enjoy writing. And I sort of agree. Writing is both intoxicating and the bane of my existence. But I'm addicted to writing qua writing, not as a mere hurdle I must jump in order to do philosophy. I fear I can never follow Mumford's method, at least not without maiming it to the point where it may no longer be recognizable as such.

Anyone who has read this far can only begin to see the extent of my tortured relationship with Mumford's method. And yet I still recommend it to anyone who will hear my advice! Stephen Yablo once called Quine's criterion "powerless," going on to clarify his judgment as not stemming from any distaste for the method. Rather, Yablo "reveres" the criterion as ontology's one last hope. With respect to writing and the Mumford method, I couldn't have said it any better myself.

[Marcus Arvan]

Hi Jon: Thanks for the kind comments!

Hi Adrian: Thanks for sharing your process! It seems to me that the Mumford method may be a great method to draft papers. However, I would be *very* surprised if it would lead to papers of a publishable quality. Judging from your comment, it looks like the method simply has you write a draft and turn it in -- without revisions. I don't know anyone who thinks they can put together something of publishable quality that way. In my experience, at least, making something publishable takes a lot of revisions. Anyway, I think there's only one way to develop one's style, and that's simply to write, and write a lot. I've found that as time goes along, my style has simply improved organically, as a result of a *ton* of practice.

[Anthony Adrian]

Marcus: You're right about the necessity of writing a lot in order to develop style. I think we need to read an awful lot, too. Reading lots of *non-philosophy* has been most helpful to me.

As far as I can tell, (Stephen) Mumford does suggest using his method to publish papers. He claims it's how he writes both articles and books. It seems to work for him!

I think I'm going to try this free-writing stuff you mention. Seems like a good idea!

[Marcus Arvan]

Anthony: I have to disagree on the reading bit. Over the course of my career, I've found that I tend to subconsciously pick up the bad habits of whomever I'm reading. When I was reading a lot of Rawls, I ended up writing a lot like Rawls (yuck!). When I read a lot of Kant, I ended up writing a lot like Kant -- something I've also noticed among Kant scholars. My experience has been that I write better the *farther* away I get from reading. This is I think an added benefit off my habit of writing out of my head, with no books or articles around. It helps me compose simply and intuitively, in my own voice.

[Pierre]

Thanks to Marcus Arvan for sharing this: although I am unsure whether I have anything like a method (which is a way to say my method is somewhat wrong), I do believe this will be helpful.

Now there remains a question that is probably of a lesser interest to those who natively speak English: as a native French speaker, even though I am a bit more at ease than many of my compatriots with reading, writing or speaking (wiz an oribel aksent) English, I still have difficulties putting my ideas into written words and adequately to develop my arguments. This has, I am afraid, the effect that what I write is far from reaching the quality I would like it to reach, not to speak of the quality expected if I am willing to publish anything (assuming it is not a plain mistake to have engaged in a PhD).

So does anyone have an idea of how I could improve my abilities in writing philosophy? Interestingly, I have a symmetrical issue when writing in French: when trying to put an idea into words, I mix French and English syntaxes and find myself somewhat confused, so that what is clear and simply expressed in English lacks intelligibility in French. (I also hope I did not make serious linguistic mistakes: this would make my whole question quite ridiculous.)

[elisa freschi]

Pierre, I completely understand your point (English is not my first language, and not even my second one). In my case, I just gave up writing in my first language, because it takes too much time and energy to develop two parallel terminologies for whatever new concept I need to introduce, and because almost all the people I want to reach do in fact read English. I even wrote my PhD thesis directly in English. I use other languages only in the classroom.

[Elisa Freschi]

By the way, I have always had the feeling that most readers of this blog come from the US and/or have English as their mother tongue. If this is not the case, show up! We might want to start a new thread about this topic.

[Pierre]

Actually, as I am at a French-speaking university (actually two due to joint PhD advising), I *must* write my thesis, as well as my class requirements, in French. Otherwise I try to write in English, but I probably lack both training (sure thing) and a good method to attain a greater fluidity. I also have very few occasions to actually speak English: perhaps speaking English on an everyday basis would help me (hypothetical yet plausible enough).

However I believe it is a good exercise to go from one language to another, as it helps me see what makes an idea clear in both languages (especially when a phrase is so written that no straightforward translation is possible). Several times it helped me understand complex issues, in part because French and English have structures so different that the extra effort on translation forces me (and plausibly other people as well) to examine precisely how a sentence, and hence an argument, is organized.

[Marcus Arvan]

Hi Pierre: Thanks for your kind comment. Unfortunately, I probably can't give you much in the way of helpful advice, as I've always been terrible at foreign languages myself (and not for lack of trying!). I guess the only piece of advice I have is the one I would also give people for whom English is their primary language: namely, keep things simple. I always figure: if something can be said clearly at all, it can be said simply, in short, concise sentences using simple words.

[Pierre]

Thanks alot: simple as it is, keeping this in mind should be helpful (and hopefully, as I “train” myself, I should become more efficient at writing).

Would you mind considering the intro of a paper that I intend to write? Getting feedback from native English speakers, and thus from people who know, on reading, what makes a paper (to begin with: its intro/motivation) fit or unfit for publication, is really difficult for me as my everyday university life is a French-speaking one.