Throw up, then clean up

I've written before a couple of times on research strategies, but my research the past couple of weeks since I began writing my book has made something crystal clear me: there is one research strategy I learned several years ago (in my final two years in graduate school) that "changed my life." Because it was such an unexpected but enormous change in strategy, one that has made the drafting of my new book go very well so far, I'd like to share it with all of, in the hope that perhaps it will make a similar difference in someone else's life.

One of the things I've mentioned before (which took some people aback) is just how fast I work. I tend to write up paper drafts in a matter of days, if not weeks. It wasn't always like this. For almost my entire philosophical career up until my last couple of years in grad school, I was very much the "slow burn" kind of worker. I would spend many, many months on end hammering out drafts, rewriting them, etc. It was hard, but I figured it was "the way" to do things.

I've made no secret that I endured some struggles as a grad student. It took me forever to come up with a good dissertation topic, but once I did, I finished the dissertation in about 8 months. How did it happen? Simple: I got a lucky break. I received an unsolicited book in my department mailbox on "how to write a dissertation." I'm usually not into self-help books, but I actually sat down and read this one. And what I read made so much sense that I had to try it. The first piece of advice, which I've recounted before, is to force oneself to write some small number of pages (say 3-5 pages) -- completely unedited -- every day you sit down to write. Although this might sound strange, every person I've passed the advice along to that has tried it has reported very positive results. I literally saw two other people I passed the book along to go from having nothing done on their dissertations to finishing in less than 9 months.

The key to this piece of advice -- as the book made clear -- is its psychological impact. One simply "feels good" when one has gotten 3-5 pages written on any given day. Once you've met that goal (if it, say, takes a couple of hours), you are then "free" to do whatever else you want that day: go back and edit, read journal articles, whatever. The point is that it gives you the feeling of moving forward every day. That feeling in turn increases your enthusiasm. You can't wait to wake up the next day to move forward more. This is exactly the opposite, in my experience, of what most grad students face (and why they often get stuck at ABD). Trying the "slow grind" style on big projects -- getting "everything right" -- leads to a negative mood feedback loop. You go home oftentimes frustrated that you didn't get much done, and it makes it that much harder to get up and do good work tomorrow. The 3-5 page a day thing creates a kind of snowball effect where you feel better and better every day, and work harder, edit better (after you finish), etc.

But I've said all of this before. What I haven't said much about is the general process it gives rise to in terms of writing: a process which (if I recall) the book I read called "Throw up, then clean up." I just can't begin to tell you how effective I think this process is.  During the past week-and-a-half, I have been writing 2,000-3,000 words on my book per day. Is it all good stuff? No! But, amongst the crap, I've gotten so much good stuff out of my head and onto paper that I can tell it's going to be pretty easy to cut out the bad stuff. And that's the thing: once you get your thoughts out there and onto paper, going back and cutting out all of the crap is *so* much easier than trying to get it all correct the first time around. Furthermore, the quicker you get it out onto paper, I find, the easier it is to discern the good and bad. I've found this way that a lot ideas I thought were good were not -- and I probably would have spend weeks or more thinking about them, whereas now I see which ideas aren't any good in a day or two. It works wonders.

So, I propose, try it out if you're having trouble getting things done. Throw up, then clean up!

Comments

[Kris McDaniel]

For what it is worth, I think this is good advice. I also try to do both of what you suggest. Get some words out of my head on to the screen. I aim for 500 words, which I think is maybe 2 pages? I also practice the vomit it out and clean it up later. Hopefully I do a good job with the cleanup!

[Michel X.]

I love your name for it, and will be appropriating it. That's exactly how I've worked for as long as I can remember, and it's just too effective to let go of it. Like you, those to whom I've passed it on have reported success. At the very least, you end the day with words on a page. They might be terribad, but it's a lot easier to go back and fix existing words than it is to spin them out of whole cloth.

The downside, of course, is that you might go through a very large number of drafts indeed. If you're not well organized electronically, that might cause havoc. I use a numbering system: X.y. X represents the draft number, and y represents the version of that draft. Whenever I open the file to work on it, I save my new work as a new file (e.g. X.y+1; so, if I started with draft 1.0 and fixed a chunk somewhere in the middle, I now save it as 1.1. When I open it up again the next day to fix a chunk at the beginning, I save that as 1.2, etc.) provided the changes are to sections/small portions of the draft. Major changes (and changes that result in what feels like a totally new draft) result in a change in draft number (X; so, for example, I might go from version 1.12 to draft 2.0). And I keep plugging away like that until I'm done.

The idea is that this way it's easy to keep track of my progress, which in turn makes it easy to refer back to thoughts I had earlier on in the process if I must. No draft/version is ever erased, which gives me free rein to experiment drastically (e.g. by making major cuts or rearrangements). If I don't like the results, I can easily re-open the previous draft/version and start afresh.

The other trick I use, when I get stuck with something (especially in the revision stages, either an argument or something more stylistic), is to open up a fresh Word page, copy/paste the offending text in there, and experiment with it as I please, secure in the knowledge that it won't affect my draft/version unless I copy/paste it back in later. Something about that fresh start really helps to get past the mental blocks, and to tackle the problem from a different angle.

[Marcus Arvan]

Michel: your organizational methods sound much better than mine. My document titles are, let's say...a bit more colorful. Here are some examples:

DraftAWFUL
DraftCRAP
DraftBETTER
DraftCRAP2

Not to mention some far more colorful things I shouldn't print here. ;) Lately, though, I've made some progress toward your way of doing things, just numbering them. And yes, like you, I *never* delete them, and always save a new document every day I finish. A lot of times I find I delete stuff that I want back later, so keeping around earlier drafts is vital!

[Marcus Arvan]

Kris: thanks for sharing!. Interesting to see how many people it seems to work for. It also seems strange to me that it's not more well-known. For instance, I don't think I've ever heard of faculty relating it to grad students. I'd be curious to see any of our readers had it related to them in grad school.

[eyeyethink]

The advice strikes me as very pragmatic, which has a good side and a possible down side. Obviously, it's very good advice if you've had trouble with writer's block or staying motivated. It's also very good advice if the aim is to write as much publishable material as possible. (And really, it's very understandable to take this as your highest priority.) However, I have doubts about whether it's the best way to produce the highest quality work you can produce.

I tend to think that good philosophy proceeds very slowly, methodically...and does not proceed to the next step until everything is in perfect order. (I'm lucky in that going slowly does not create many brain-cramps or sap motivation.) The goal, I think, is to be _as clear and as convincing as humanly possible_. I don't think the profession supports working in this way, but the work the profession supports is not necessarily the best work.

So its a question of whether you want/need to be more productive, publishing-wise, or more methodical. I tend to favor the latter, to the detriment of my publishing record. Which is in some ways irrational, I realize.

[Christopher Stephens]

I got this same advice indirectly from an Arizona prof (don't know if it was the same one you got it from - D. Schmidtz in my case) - who recommended two books - Writing with Power (Elbow) and another one too (maybe How to Complete and Survive a Doctoral Dissertation). But my current favorite that I recommend to all graduate students I come into contact with is Professors as Writers, by Robert Boice, which gives the same basic advice (write regularly for a set time or number of words and keep track of your progress (or lack thereof). How to Write a Lot (by Paul Silvia) is also fun with the same basic advice.

Works wonders.

Incidentally, regarding eyethink's concern, I think there is empirical evidence discussed in one of these books that those who follow these strategies of writing regularly report having more and better ideas. So there's a case to be made that it improves quality as well.

If you're really concerned only to publish once you're super duper confident in the quality and clarity of your papers, you can still do that following this strategy. It doesn't say you have to publish all the stuff you write. But the evidence suggests that this sort of regular writing improves the quality and clarity of your thinking, and gives you more ideas.

[Marcus Arvan]

Chris: I wouldn't be surprised at all if there were empirical research like that. My experience is that I have become a far better philosopher -- with far more and better ideas, and a better writer -- as a result.

eyeeyethink: If I recall, you've expressed these sentiments before. But here's my question for you. Have you ever actually *tried* it? It's one thing to say "slower is better" if you've actually given both approaches a serious go. But, if you're just defending "slow is better" because it *seems* that way to you, aren't you just being dogmatic? As I note in my reply to Chris, I sincerely believe I have become a far, far better philosopher since I began this practice several years ago. My fear -- and I mean this in the kindest way -- is that you are avoiding trying something that would make you *better* out of some kind of abstract belief in your own process. If the slow thing really works for you, that's all fine and well. But as you note yourself, you've been having trouble publishing. What if -- as I suspect -- you would actually do better, and more work, by trying the "throw up, then clean up" strategy? Wouldn't it be, as you say, irrational not to give it a shot?

[Michel X.]

eyeyethink: I don't think the vomit+cleanup method is necessarily any less methodical. It all depends on how much time you spend cleaning up, really. I don't think there's any reason to think that a paper that's gone through, e.g., forty drafts in 40 days is any less good/methodical than one that's been through one or two drafts in that time period (if anything, I suspect the reverse is true!).

I think that the significant difference between the two methods is that with this one you can easily see where you've come from; it leaves nothing to the imagination.

[eyeeyethink]

Maybe I am being somewhat dogmatic. Though not entirely--in this profession, sometimes you *must* write something quick, e.g., comments on a conference paper. And in my case, the quickly written comments tend to be worse than the slowly written comments. However, maybe I'm failing to distinguish time spent on a first draft from *total time* spent on a paper. Michael is right that clean up can be very methodical, so I stand corrected when that is the case.

[Marcus Arvan]

eyeyethink: I have a feeling that your experience writing conference comments quickly -- and how writing them quickly compromises their quality -- is colored by a couple things.

First, as your comments suggest, you seem to be equating "writing a draft quickly" with not spending a great deal of total time on the paper. Look, I've dashed out conference comments too, and they're sometimes not very good. But this is *not* because I wrote a draft of them quickly. It's because, after writing them quickly, I didn't go back and do all of the hard revising work. So, it's no surprise to me that your experience "writing quickly" has been that it affects quality. I am not just advocating "throwing up." The maxim is throw up, then *clean* up!

I also suspect that there may be some confirmation bias at work with you. Suppose someone tries a (bad) whiskey once or twice and then concludes that whiskey is terrible, confirming their prior belief that whiskey is terrible. This person's sample size is too small, and they've given up the moment their few samples confirmed their preconceptions. I wonder whether something similar is going on in your case. You've tried writing quickly a few times, and you see that what comes out is word vomit...so you go back to your normal process. But the whole point of the process I am advocating is to stick it out! Pretty soon, you'll start cleaning up the vomit, and moreover, the more you practice writing fast, the less vomitous your first drafts will be.

In other words, I mean to say: don't knock it until you *really* try it over a period of weeks straight. Given your self-professed issues not publishing more, what do you have to lose? (Note: it was desperation in my own case that led me to give it a shot. I didn't expect it to work, but once I really followed the advice -- all of it -- it shocked me how much it improved both my work and my enjoyment in doing it. I really do believe that if you give it a serious go, chances are you'll surprise yourself! Again, I don't know a single person I've passed the advice onto who didn't report very positive results).