The Mechanics of Your Research Production
There are many little everyday steps leading up to the production of a philosophical manuscript. Charles Rathkopf (CUNY) writes in asking philosophers about “the daily habits of routine research.”
Do they take notes on paper, then transfer to a computer? Does anyone still try to write anything substantial with pen and paper? What bibliographic system is best? Do people keep pdfs of the papers they read on their computer? Or just save the references? Do people use comment functions in pdf readers? Do they keep physical copies of the papers they read? What software do people like to use for note-taking? For writing articles? How do people deal with logical notation?
Some of these questions have been addressed here at Daily Nous, with posts about writing and reference software (and just a few week’s into DN’s existence we conducted a little poll about when people write). But people are welcome to revisit these questions and the others Rathkopf lists, and we can add further ones that might be of interest:
Do you keep files of notes for paper ideas, and how do you organize them? Where do you write? What kind of devices do you write on? Do you write only when you have long stretches of time available (how long?) or are you able to squeeze in writing whenever you have a few minutes to spare? Do your ideas for papers arise mainly from conversations with others or when you are by yourself, reading or thinking? How do you decide which of many ideas to try to turn into papers? Do you stop writing when you get stuck, or do you leave your document knowing what you are going to write next, so it is easier to start again, or write for a fixed amount of time each day? Do you time your work? Do you turn off email and social media while writing? Do you take naps to recharge?
Further questions, along with answers to any of them, are welcome.
Short answer: I have no idea. But here is some of the things I do. I read exclusively on an iPad, typically using a PDF reader. I annotate by highlighting and write down comments when they occur to me. Sometimes this reading triggers ideas – which may be more or less closely related to the topic of what I’m reading – that turn into papers. When I’m excited enough about an idea, I begin to read things they seem likely to be useful in developing it (again annotating PDFs). If I’m sure I want to pursue the idea after some reading, i ususally go through the annotated PDFs, writing down my own comments and situating them in the context of the claims of the paper. I do this on my laptop. I then write the paper, with iPad beside to refer back to the papers and books, and with my notes documents open. I write the introduction to the paper, setting out the debate and putting forward my argument (“I will claim that….”). At this stage, I have no real grasp of whether the idea will really work. I write a first draft which is as complete and well structured as possible. All going well, that first daft just needs polishing before it is submitted to a journal. All does not go well on many occasions. Sometimes I discover the idea is more or less hopeless, or not very interesting once suitably qualified. Sometimes I realize, once the first draft is written, that it needs to be entirely restructured or that the central idea is hopeless but one that had seem peripheral when it emerged in writing is interesting enough to be the focus. I do not edit the first draft then; rather, I start again. I am often able to cut and paste some paragraphs from the first draft into the new document as I go.Report
Scrivener for book writing, then export to Word, adding references from Scrivener notes with Endnote.Report
IThought has been changing how I work in a very positive way.http://toketaware.comReport
The main thing that I do is write for at least 90 minutes first thing every morning. I actually get up a 5 AM to do this. Once I get an idea about what I want to write about, I start by creating a word file that I use to write down my thoughts in a non-committal way. In this document, I’ll create an outline of how I might structure the paper, record possible titles, make lists of what I need to read, provide definitions for any terminology that may prove useful, come up with examples or cases that I may want to discuss, record my thoughts and criticisms on what I read on the topic, construct sketches of main theses and my arguments for them, write various paragraphs that I think that I may need, etc. This document is used for brainstorming, and I find it very easy to write in this non-committal way. Once I think that I got all the basics worked out (thesis, argument, organization, etc.), then I create a new word file that will be the paper itself. The paper is written, then, by cutting, pasting, polishing, and further developing what was in the brainstorming document. Of course, a lot of what’s in the brainstorming argument turns out to be useless and just remain there, never making its way into the paper. For citations, I create snippets in TextExpander for the complete biographical details of everything that I ever cite. That way, the next time I cite that paper or book, I can just use the snippet to expand it in a footnote or bibliography. For instance, “pportmore2011” expands into “Portmore, D. W. (2011). Commonsense Consequentialism: Wherein morality meets rationality. New York: Oxford University Press.” I keep PDFs everything that I read, including books if I can find them. This allows me to access them whether I’m on the road or in a coffee shop.Report
Oh, like Aaron, I also sometimes use iThought at the very early stages.Report
I actually really like footnotes and citations; it can be hard to keep two different systems straight (e.g., Chicago for philosophy, Bluebook for law), but that’s part of the fun. I’ve played around with some of the reference tools, but I think they make me enjoy writing less. Writing out the citations helps me “rest” mentally, and it’s fun to revel in some of those technical details without assistance. (Of course I don’t think either approach is better or worse, just personal preference.) I also rarely write from notes or outlines; I spend months writing and rewriting in my head and, when I finally sit down to do it, the article’s mostly done. I’ve noticed a lot of colleagues spend more time thinking while writing whereas others spend more time thinking before writing. I have friends turn in absolutely terrible first drafts that eventually morph into spectacular papers, but the progression between my own drafts and final versions is far less ambitious–which either says my drafts are good or the final papers are bad. 😉Report
Recently I have been working on a book manuscript, writing a set number of words per day usually on an Alphasmart (see http://en.wikipedia.org/wiki/AlphaSmart), which is a sort of digital typewriter that can transfer text to my computer using a USB cable. It’s a wonderful way to write while avoiding distraction. I then put the text into Scrivener (http://www.literatureandlatte.com/scrivener.php), which is a program designed for novelists where your master document comprises a bunch of shorter subdocuments (and sub-subdocuments, etc.) that you can easily drag and drop to reorder, duplicate to revise, and more. It’s been a very nice approach so far, though we will have to see if the ideas are any good.Report
Douglas Portmore’s comment was incredibly helpful! As someone trying to develop a writing process as I work through my dissertation, I would especially welcome more reports of the kind that he offered. For those of us still trying to iron out the nuts and bolts of how to get writing done, such detailed descriptions of how successful philosophers do it are a godsend. I tried incorporating some of what Dr. Portmore described when I sat down to do some writing today and it helped me get out of a rut that I’ve been in for several days. Thanks so much for your post Dr. Portmore (other posts have been helpful too)! I hope that more philosophers will consider describing the details of their writing process in order to help us young philosophers out.
(As an aside: I wish that my grad program had done a little more to help grad students develop better writing practices – I suggest this as a topic for those charged with running professional development courses / workshops!)Report
From notes in the margin of an article I’ve printed out or a spare sheet of paper, to a conversation standing around a whiteboard (if one is available), to LyX or a LaTeX editor, to a draft printout to be filled with red pen scrawls, back to LyX or LaTeX. I can’t read anything serious on the screen, and I hate seeing my work poorly typeset even in its earliest drafts—it’s just distracting. I do all my footnotes and bibliography by hand, because no BibTeX package precisely matches the conventions I’ve developed over the years. I also always add full bibliographical details as I go along: it gives my mind a chance to take a break on a regular basis, and it avoids my having to go through the whole thing at the end to fix citations.Report
@Beau This is slightly derailing, but one of the greatest perks of a LaTeX workflow (imho) is the ability to use BibTeX or BibLaTeX to automatically manage your citations. If the issue is that no existing BibTeX style file (.bst) suits your needs, I would suggest creating your own. Most LaTeX distributions come with a command-line utility called makebst that allows you to generate your own .bst file based on your answers to a number of questions. The utility is also available from CTAN (http://www.ctan.org/tex-archive/macros/latex/contrib/custom-bib/).
Now for the actual topic of the thread, I’m a fan of the following technical set-up, at least. I use LaTeX primarily for the nice typesetting, unparalleled mathematics support, and BibTeX integration. I make sure to put each sentence on its own line and use a version control system (git) to manage versions of my papers—one version that’s a chapter in my dissertation, versions for each of the dozen conferences I might have submitted to, versions to reflect changes made after receiving comments, etc. This allows me to revert to previous versions without having some cluttered folder with all these different files. I can also diff the files to see changes highlighted (this is where separate lines for sentences is important; changes are shown by line and so a sentence per line is a nice way to make sure changes are tracked in reasonable units).
For taking notes I like to annotate PDFs, though I haven’t found a single editor that didn’t strike me as clunky in this regard. I annotate everything as I’m reading it (when I’m being good at least…) and sometimes the various comments turn into papers. Generally I develop ideas for papers on my own while reading (or driving, I find that driving home after teaching is often a productive time for me) but often I will “try ideas out” in conversation before committing substantial effort to developing them.Report
Of course I’m still settling on a workflow, since I’m just starting out, but I’m much happier with where I am now than I was a year ago. The decision that has helped me most is to totally divorce authoring/content generation from publishing (in the sense of making a final product for public viewing). Content generation is done all in markdown (I like Editorial on the iPad and iA Writer on the computer). That’s where I sketch ideas, where I draft, where I make notes to myself to fill in bits later. The markdown files sit in folders organized by project on Dropbox so I can access them across devices. At any stage when I want to show my work to someone, I use Pandoc to convert to TeX. I have a standard header that will generate the typesetting I prefer, and BibTeX takes care of the citations if I’m at a stage where I’ve filled them in.
The reason I’ve found myself liking this particular setup is that it allows me to do the creative work in a completely distraction-free environment. Plain white screen; plain text; minimal markup. If I even have the option of futzing with spacing or formatting I’ll lose time to it (I don’t want to think about how much time I spent over the years yelling at a computer screen “Word, why are you doing that?!?! Stop being helpful!”), so I remove the option.
Pandoc is actually capable of handling chunks of TeX, so if I have more time to write than I have energy (rare, but it happens!) I go through and replace “CITE so-and-so” with the correct citation markup or “PHI AND PSI BOXARROW BLAH” with the correct math markup because those are pretty mindless tasks. But I try not to do those things until as late as possible, since I want my authoring environment to be as clean and plain as possible.
Versioning is pretty clunky for me. I keep major drafts (usually ones that went out somewhere) organized just by a date in the filename, and then a scratch file for each project where I put all of the large sections I decided to cut but couldn’t part with. Git just feels like overkill for what I’m doing, but maybe I’ll feel differently later.
For PDF annotation I like iAnnotate PDF. Keep those on Dropbox as well so they’re easily accessible both from tablet and computer. I read better on paper, but can’t resist how much of an advantage it is to have all of your notes and sources searchable.Report
My own approach is a hybrid. I read on paper – often printing PDFs of articles I find through research databases – then take notes longhand. I then transcribe my hand-written notes into a Word file and attach it to the EndNote reference to which is also attached the PDF. I also print the transcribed notes and keep them in the file folder for a particular writing project.
I discovered last summer that, when I need to draft a paper or other longer, formal piece of writing, I really do best with a legal pad, a couple of fountain pens, a folder full of notes, and at least a quarter mile between myself and the nearest piece of communications technology.
On a couple of occasions, I have taken myself “hostage” by leaving my phone and computer at home, walking to a local coffee shop, and not allowing myself to leave until I had written a particular number of words.
Two sessions like that produced the draft of a paper, which I then transcribed into a Word document with surprisingly little revision.
When I write for my blog, I tend to compose on the computer, except when I don’t.Report
I’ve found Peter Vallentyne’s advice for writing dissertations to contain helpful advice for writing generally. See the link “Writing Strategies”, at
I approach the task of research with the same intensity of focus that other professionals apply to their respective industries. I set my alarm to go off very early in the morning, and then I do something almost unthinkable: I wake up one full hour before it ever goes off. This represents a struggle since it requires me to issue myself the challenge while still enjoying the delicate subconscious offerings of my final REM cycle, but it’s worth it just to sit up on my bed, lean my head to the side, and stare down my alarm clock with the smugness of Kobe hitting a game winner on infinite loop. This really sets up my day. Was it not Whitman who applauded “the song of me rising and meeting the sun”?
Once I’m out of bed, I head over to my computer to begin my research. What do you think I do first? Do I open a new browser window so that I can have an array of research sources optimized across neatly divided tabs? That’s exactly right. But it’s what I do next that will surprise you.
What I do next is call my mother and tell her that I won’t be coming over later for dinner. A tear nearly trickles down as she informs me that she’s making lasagna. If I could just touch the hem of Aquinas’ garment, I reason to myself, a wave of perfectly debilitating asceticism would wash over me whole. The Summa is nearby. I grab it and find a note inside that I had left for myself the night before. It reads: “You’ve wasted enough time this morning. Get to work.”
You wouldn’t believe the jolt of electricity that this note manages to produce up and down my spine. I’ll tell you this: it’s enough for me to have not only an internet browser window open, but the word processor now open also.
What happens next is, to borrow a motif from the works of J. K. Rowling in her series on Harry Potter, a shy but fiercely loyal and comically unassuming young wizard predictably thrust headfirst into a war of good versus evil, magical. The letters begin spilling onto the page. An argument here. A thought experiment there. Hours pass. Days go by. Weeks bleed into each other. Do I visit my family on Thanksgiving? I don’t have time. Christmas shopping? Let’s have Santa Claus take a cruise on the old Amazon.com river, shall we? No sense in making the shopping mall rounds when my browser window is already open from all the research I’m doing.
I only stop to experience the world again when a complete book is finished. The publishers will love it. I even have the dedication section done. To my mom, whose lasagna went cold and uneaten so that this world-healing scroll might be written.
Once the manuscript is approved, I download it onto a thumb drive, walk outside, and toss it way up into the sky. The imagery here is dripping with metaphorical sizzle. What I’m trying to say is that I upload the file into the cloud (LOL now do you get it?) and then give myself the rest of the evening off. The words of Yeezus are most appropriate here: “I always had a passion for flashin’ / Before it happened, I closed my eyes and imagined.” I just close my eyes, envision my next research project, and as I drift off into unconsciousness I let myself smile a little. The alarm clock will ring soon, and I have to wake up before it does.Report
I have thought and written about my philosophical research workflow and production in some detail on my personal blog. I link to those posts here in the hope that others might find them useful:
For the technical setup, I read pdfs with a nice free app called Skim, which is great for marking up and commenting on articles. These articles I store in a BibTeX database with linked files. After reading the paper, I have a script that converts all highlights and markups into text (I use evernote, but it can do it to .doc too), with all the bibliographical data. The highlights/markups also show up in the BibTeX database when I click on an item just to see quickly what it was.
After this, for the actual writing, it really depends. If it’s a longer project, I often use Scrivener, which has nice outlining features, but with the LaTeX code (and then paste it to the Latex editor). If it is shorter, or in the editing stage already, then just edit the latex script.
All in all, I can’t recommend BibTeX highly enough — it is free, makes inserting references so easy, and it even organizes your pdf files for you if you want it to. I keep my whole BibTeX library on Dropbox so that it’s accessible everywhere, with all the files organized by BT. And if you use Skim, it integrates really well with that too.
Otherwise, I’m also a fan of writing in the morning — 2 hours for me usually, and I only cancel it in cases when I would cancel teaching. No excuses. Even if I’m tired and all my writing is crappy on that day, it just makes it so much easier to think and do the rest.Report
Correction: BibDesk for BibTeX organization/setup.Report