Tuesday, June 4, 2013

[Forgot to Title This Post]

In comments over here, anonymous 8:56 writes:

Hey Smokers, 
I am beginning the first phase of my dissertation research (the topical), and I was wondering if anyone had any specific tips about how to keep track and organize articles, reading notes, etc. and perhaps any tips on how to go about working through a large body of literature. Do you take notes on every article? If so, how much time do you spend summarizing an article? How many articles/chapters do you read per day?  
Thus far, I have been logging everything in a word document, but it's getting out of control...I really need a better (and hopefully less stressful) system 
I really appreciate any advice.

If there's one thing I've learned from the various discussions of research and writing strategies on this blog and others, it's this: there's no one way. I've got things that work ok for me and various anonymouses have things that work for them, but there's nothing anyone can tell you that will be guaranteed to work. Moreover, the dissertation is a learning experience. You're supposed to struggle with it. I know you know all this, but it bears emphasizing. If you knew what you were doing, it wouldn't be a learning experience to do it.

When I'm in reading/research mode, I generally don't compile a formal bibliography or anything like that. Annotated or otherwise. I make use of a couple of bibliography/PDF organization applications, BibDesk and Mendeley (I don't really like Mendeley; I use it because it's free), and I use keywords to help stay organized. But that's about it.

Then I just sort of read, at whatever pace feels right. I start by taking notes in the document itself, be it a book, JSTOR printout, or PDF document, and I start taking more detailed notes in a separate medium only if I can see that the article (or book or whatever) is going to end up being important. But I don't try to read a certain number of things—articles, pages, whatever—per day. I try to give each thing the attention it deserves; sometimes that means breezing through, and sometimes that means spending a week or more on one article.

When I'm dealing with an important article, book, or chapter, I open a .tex file and take notes in it. Depending on where I am in the project, this might be a separate file devoted just to this one article (or whatever), or it might be section in a larger document. Here, I try to articulate the views and the arguments, how the author defends the premises, how the author responds to criticisms, how the material fits into the larger picture, and my own reactions to the material. At this stage I am meticulous about documentation. I quote passages that support my interpretation and cite page numbers. Always cite page numbers. Always. I think about how the material should be organized—I think about what order things should go in. Often, this is not the order in which it appears in the source text. I find that I end up spending a lot of time moving stuff around—I find that I struggle with organizing the ideas more than almost anything else.

At this stage, the distinction between “taking notes” and “drafting” is pretty thin. A lot of the time, I'm taking notes in the actual document I'm writing. However, I still try to be careful in distinguishing between what we might think of as “notes” and what we might think of as “writing”—that is, the very rough “drafty” stuff and the semi-polished stuff that I'm more-or-less satisfied with. (Of course, everything is subject to revision, but some things are more subject than others.) Here I find LaTeX's percent signs to be very valuable. Drafty stuff gets a percent sign with a [bracketed label] indicating why it's percentaged. When the passage is ready for promotion, I delete the percent sign and a new paragraph is born. This also lets me excise material in a slow, noncommittal way. I find that those percent signs get a lot of use.

So, to get back to 8:56's specific questions: I take notes on every article, but often just in the article itself and not necessarily in a separate medium; the amount of time I spend on an article depends on its importance for my project—although I probably wouldn't spend much time summarizing the article, exactly, but would spend whatever time was necessary to summarize the particular arguments or views or whatever that were pertinent; the number of articles/chapters I read in a day depends on the articles and chapters. I don't try to keep summaries of everything in one document; I don't even try to keep summaries of everything.

A dissertation is obviously a big project, and it's obviously going to be difficult to keep track of all your research. So I wouldn't try to keep track of it all in the same place. I'd carve the material up into manageable bits. It is, obviously, customary to organize one's dissertation into chapters, so I'd start there. Maybe organize into sections or subsections if there's a particular chapter that's getting unwieldy. But I suspect that the kind of annotated bibliography 8:56 alludes to, containing all your research for your dissertation, is going to be more trouble than it's worth and will cause you to spend more time than you should on articles and chapters that aren't central enough to your project to be worth it. And, as 8:56 mentions, it will be hard to keep a document like that organized, and so the document wouldn't even be particularly useful.

Anyways, that's what I think. What do you think?

--Mr. Zero


Molly said...

You need to get a citation manager program (Refworks which is subscription based through university libraries) or Mendeley which is open source and free. And, see if your university library offers classes on dissertation management and research.

Anonymous said...

Thanks for this. I think it's important for people to hear that what works for someone might not work at all for someone else, and that the dissertation process is a learning process. You will mess up as you go along. You will have regrets. You will do it better the next time. I think my adviser might have summed it up best when I asked him about this years ago (this was in regard to figuring out how to schedule writing, but I think it applies more generally), "I don't care what kind of system you use just find something that works and do it!" People are frequently way too quick to assume that what works for them is *the* way and that it will work for everyone else (and any other method will ultimately fail). Trial and error. You'll probably never write a good paper without at least a few drafts. It's extremely unlikely that you can write a good dissertation without screwing something up somewhere along the way.

Anonymous said...

I just use a private blog to summarize parts of the article important to my own project. I might jot down page numbers to return to but nothing terribly involved. Most secondary work is either not very good or not especially helpful for my own project. The blog is useful because I can add tags, change or remove posts, update items and even organize several blogs for different projects.

Anonymous said...

I've tried different systems but what has ended up working best for me is to (1) annotate pdfs for articles and (2) keep notes in text documents, one for each topic of interest (e.g. GameTheoryNotes.txt or DavidsonNotes.txt, etc). Later those notes and annotations end up helping quite a bit. I revise them a bit every time I look at them. Often the note files go through fissions or fusions and they also contain bits of text I can paste in to papers. I agree that there is no one system for everyone. That's why this kind of blog is useful. You can scan a bunch of other people's suggestions and try things out. Oh for getting a feel for a literature I try to find good reviewish articles (e.g. SEP / phil compass) to start. Sometimes I find someone to read and discuss a paper with since it forces me to actually read something I don't want to read. I sometimes spend multiple days on important widely cited stuff and for less relevant things just read the abstract and scan. It takes me a long time but maybe I'm slower than most.

zombie said...

Highlight, marginal notes, stars, underline, multiple colors for the really important stuff. I still need paper for the main documents I use, but all of this stuff can be done digitally too.

On the most recent paper I worked on, I found my articles to be completely unmanageable. They number over 2,000 at this point (not for the single paper, mind you, but overall), and I had them roughly organized into topical folders. I just couldn't find anything, and it got ridiculous. On top of that, my institution has kind of a crappy library, so I have to order most things from ILL, so it was putting a serious crimp in my research to not be able to find stuff I knew I already had.
So I finally sprang for Papers, which is an archiving program. It's a little labor intensive at the moment, but that's because I'm importing a couple thousand articles. If you do them one at a time as you download them, not so bad. Takes a couple of minutes to keyword. It's pretty smart about author names, and for many true pdfs, can grab the info automatically (sadly, not for the crappy xerox pdfs I get from ILL -- another reason I have to print things). There's also a notes field where I can put info like what paper I cited it in, or whatever. It has a citation manager, although I haven't used it yet since I was halfway through my paper when I started.
Not free. I think it was $79 (30 day free trial). There may be free programs that do the same thing, but this one worked for me.

I've poked around at Scrivener a bit, and it looks pretty good. It's a word processor/editor/project manager program, which includes a citation manager. I still have not found a word processor I like for academic papers, and I tried 3 or 4 in the course of writing my diss.

One suggestion: decide on a citation style at the beginning, and use it consistently throughout.

Anonymous said...

Anon 8:56 from the original post here.

Thanks so much guys, this is super helpful. I'm glad to hear everyone stress that this is a learning process, and that there isn't a *correct* way of dissertating!

I am going to test Mendeley and try to get my hands on Refworks.

Anonymous said...

I use scrivener to do the writing. I then use papers to store all my PDFs on my iPad ( it syncs with papers installed on your computer).

Wonderful combination.

Anonymous said...

I'm still dissertating (and still pretty early on), but like Mr. Zero, I think that having all my notes typed out or in a single document is a waste of time.

I've got my articles in topic-differentiated folders. I highlight and annotate them on the screen (and a few on the page, which I then scan and send to myself to file away). This way, it's easy to make sure everything is backed up. My books are in a separate book folder when they're electronic. When they aren't, I *do* take the time to type up my notes after I've annotated stuff on the page. That way I get a refresher and can easily back them up (plus, I can return the books to the library). When I work on a section, it's sometimes helpful to copy/paste some notes or highlighted text directly into a Word document to start; when I do that, I'll save a copy as a "notes" file. Otherwise, I just refer back to the articles themselves, since my notes are there.

It may not be the single most efficient way of doing things, but it makes things pretty smooth for me.

I've got ReadCube for citation managing. It's ok, but I've not really invested the required time into learning (or using) it yet.

Anonymous said...

I use zotero for citation management. It's free and easy. You go to google scholar, find the reference, press a button, and it's in your library. It saved me so much time on one project I donated $10 to the project and they (well, gmu's fundraising dept.) have probably sent me $20 worth of printed glossy promotional spam since then (despite my efforts to stop them--donate anonymously!). But I have not seriously researched this. Are the other citation manager program mentioned here better than zotero for some reason?

Anonymous said...

I want to second the mention of Scrivener. It's not got much built in for bibliographic management, but it's the best writing program I've ever used. It is useful all the way through a project: it helps me organize notes from books and annotated pdf's, and then make use of them when I'm writing. I'm currently near the end of my dissertation, and the way Scrivener makes it easy for me split my writing up into small pieces has helped my sanity.

Xenophon said...

Two quick comments that relate to the next stages of your dissertating: (1) (which Mr. Zero already mentioned) don't fix a firm divide between research and writing. Start writing as early as possible, preferably today. There's no end to the research you could do in a field like philosophy. (2) Once you start writing, back up regularly. How regularly? Ask yourself how much work you're willing to lose, be a day or a week or a month. Use an external hard drive, but also use the cloud to make sure you're protected against acts of God that take out not only your computer, but your apartment and your university office.

Good luck and have fun. I loved writing my dissertation. It was the first research project I did that was all my own.

Anonymous said...

For note-taking I use Evernote, which is free and multi-platform, so you can use it on different machines and OS's. E.g., I have it installed on my PC notebook and iPad. And it syncs and stores everything on the cloud, and you can sort by folder and tags.

MS OneNote probably is a better note-taking program, but its functionality is limited in its free iPad version.

I store PDFs on Dropbox and Sugarsync (and there are lots of other backup and syncing services where you can get gigabytes of storage for free), and use free apps to sync them to iPad 4 where I can read them on its retina screen (very clear, not at all tiring for your eyes, and you can read them in the bed, on the pot, on a subway train, whatever). Important bits of text in your PDFs you can copy-and-paste into Evernote or (if the PDF is of images rather than text) take a snapshot of the viewable screen by pressing the Power + Home buttons simultaneously.

Anonymous said...

I'll third the Scrivener recommendation. It's a great program for organizing a big project, and has some tools that can help with keeping track of research. Certainly worth the free trial.

Alan said...

Great comments, and I hadn't thought how much the Cloud might be important for backup until the thread mentioned it. And made me think of the darkest night of my grad career, before completing my PhD at UT-Knoxville. I was finishing my MA in 79--proud of it--I actually had already published something out of it--when I was picking up the finished copy from my paid typist. I put it on the back of my Honda 350, confidently double-strapped with bungee cords, and rode home. And there, stopped at a stoplight on Kingston Pike, I looked back--to see my finished MA thesis wafting among countless cars in the breeze of Knoxville rush-hour traffic, crushed under tires and slipping onto the sidewalk. I cried. Then went to a celebration of my completion and drank not a few drinks. The typist gave me a price break on the retype, bless her heart.

Anonymous said...

Here are some things that I found useful:

1) After you finish taking notes for a paper, read them immediately. Studies show it helps remembering. It means budgeting some time to go over your notes at the end of the reading.

2) I do take all my notes in a separate notebook. I like using big sketch books. I have one sketch book per "topic". I summarize the text I read in black, and I write my own thoughts in some other color. I also write down at the bottom of each page references from the paper I want to read at some later time.

3) I'm actually not a huge fan of starting to write too early. Of course, I start writing my thoughts immediately, but I don't start writing a more elaborated and structured text before I have a better sense of the bigger picture.

4) But I do take the writing process to be part of the thinking, not just a putting-down-on-paper of already-figured-out ideas. I find it magical. And somehow this thought makes writing less scary for me.

Anonymous said...

Does anybody have any advice for putting the whole thing together (i.e. chapters, table of contents, front matter, etc.)? I was thinking of writing each chapter in a separate document and then somehow compiling the whole thing at the end, but I have no idea how I would do that. Would Scrivener help? It's pretty expensive.

zombie said...

I wrote each chapter (including references) as a separate document. Some of my chapters were pretty long (60+ pages). It gets pretty uwieldy (IMO) to try to work with hundreds of pages in a single document. Once it's all done, you can repaginate each chapter, starting subsequent chapters on the correct page number. It is easy to merge a whole bunch of documents as a pdf.

Having independent chapters also helps if you're feeding them to your advisor one at a time for feedback.

zombie said...

I second using Sugarsync to store docs and move them from one machine to the other. I use it all the time to put papers I need to read on my iPad. It's free, and they give you ample storage space. On the iPad, you can read them right in Sugarsync, or move them to a pdf reader like Adobe PDF Reader.

Recently started using Google Docs to store papers in the cloud, so I could offload them from my computer. Also free, very easy to use.

Anonymous said...

Non-Computer Advice:
Most of the comments here emphasize how much reading is involved in writing a dissertation. I agree you could read a ton.

But don't try to read everything before writing. Many students I know end up writing long book reports for dissertations precisely because they see the dissertation as a research catalog of their readings rather than a piece of research.

Computer Stuff:
1. I second the use of Zotero. It's also platform independent if you happen to use, say, a mac at home, a pc at the office, and a linux machine at your friend's place.

2. It's easy to write separate chapters and compile them into a single document using Latex. It also looks nice and is free.

If you use a nice Latex editor (Texmaker or WinEdt, for instance), you can automatically add citations that you've added to Zotero. TexMaker is free, but WinEdt, I believe, is not.

6:38 said...


Thanks for your help! Did you merge all the documents as a .pdf using MS Word or some other software?

zombie said...

I used several word processors when I wrote my diss. I kept trying new ones. In the end, I used Pages (Mac), but it doesn't have anything Word doesn't have. Within Pages (or any decent wp) you can repaginate, starting each chapter on a page number you select. Adobe Acrobat makes it really easy to merge a bunch of separate docs into one, in the desired order. (I also used it when on the job market, to merge separate dossier elements into a single pdf for emailing.)

If you don't have Acrobat, you can always copy-paste within a wp to create the merged document.

6:38 said...


Thanks so much for that. I now realize that that kind of information shouldn't have been to difficult to track down, but I'm so bad with technology that I just had no clue where to start. You've helped to relieve a great amount of stress and anxiety.

Anonymous said...

Here's the system I've been using ever since I started my dissertation, and that's still working for me now 10 years later. Whenever I start reading a book or article, I open up a word document and title it with the author and title of the book/article (e.g.: Lewis - New Work for a Theory of Universals). Then I keep a running tab of what happens in each paragraph. Just one line of text per paragraph, rarely a complete sentence, and I don’t worry too much about the accuracy or completeness of the characterization. I'll also insert my thoughts about what I'm reading at the relevant points. So it ends up looking something like this.

8a Against Armstrong's defense of universals
8b Against indispensability arguments
-Is he just rejecting using indispensability arguments to rule out/in universals, or is he objecting to indispensability arguments in general?
9a An ontology of possibilia; properties = sets of possibilia
9b On whether universals should replace possibilia
9c Outline of the paper

Universals and Properties

Also, I'll highlight important paragraphs and important notes-to-self by putting the line in bold font.

I keep all these documents in a single folder "Books and Articles", which also contains pdfs of the articles themselves (labeled the same way).

I've found that taking these notes helps me stay focused when I read, and the notes can be extremely useful several years down the road, when I vaguely remember that Lewis said X somewhere in his paper, and I can just pull up the notes and skim through them (or do a keyword search) to figure out where exactly he said it.

Anonymous said...

Anon 7:56am:

That is an excellent tip! Thanks!!!

I am going to try a mixed approach: rather than summarize each para, use paragraph markers to indicate where new lines of argument or bits of exposition begin and end. Would that work just as well?

(Am I weird for finding this post more exciting than the McGinn ones?)