Tuesday, May 04, 2010

256 Claims

How many things do you need to know to write your PhD dissertation? Well, it varies from field to field and program to program, but here's one way to estimate it. How many prose paragraphs are there in a dissertation? Each paragraph will consist of one key sentence and about six sentences that elaborate or support it. A paragraph claims that you know something (expressed in the key sentence) and tells the reader how you know it (in the remaining five or six sentences). That is, each paragraph presents (exactly) one thing you know.

So how many paragraphs? With one or two paragraphs to the page, 250 paragraphs seems like a good ball-park figure to begin with. In one sense, then, your dissertation is a presentation of 250 things you know. All of us know much more than that, of course, and you will know literally countless things in your area of specialization before you are finished. You won't claim to know each of them specifically in your dissertation. Your dissertation will be a selection of 250 things you know well.

Not only are these things merely the tip of your mental iceberg, you will be able to present only the tip of your knowledge of each of them. That is, after three years of study, you will be able to offer much more than six sentences of support for each of the claims you will make in your dissertation. Your challenge as a writer is to select, from among these (again, countless) sentences you could write, the sentences you must write to convince your reader.

The 250 knowledge claims that make up your dissertation are not about isolated facts, of course. They form a system. That means that they can be grouped and ordered into hierarchies. Every tenth claim, more or less, is a summary of claims at a lower level of aggregation. 225 claims at the ground level. 25 claims the next level up. Probably five claims at the next level. And then one claim that summarizes your whole dissertation. 256 altogether.

Now, here's something interesting. Imagine your dissertation as the defense of 256 claims, organized roughly as I just suggested. What would those claims be? Some will be about what you did (method). Some will be about what you and your field knows in a general way (theory). Some will be about your object as such (results and analysis). Some will about the implications of your work for practice, whether social, political, or managerial (recommendations). And some will about you and your dissertation itself (introduction, conclusion, preface). Like I say, about 31 of them will merely summarize and generalize the claims of the other 225.

Now, imagine spending exactly one hour trying to write each of those 225 substantial claims at the lowest level of generalization. Spend half an hour on the 25 + 5 summaries. Leave that 1 overarching claim, which is your "thesis" proper, out of the plan for now. Do the math. 240 hours. Give yourself 5 three-hour sessions per week (writing from, say, 9-12, Monday to Friday). That's 15 hours per week. 240 divided by 15 is 16. It will take you 16 weeks to write those 255 paragraphs. All you need is to stick to a schedule and have that list of 255 claims that, taken together, are your bid to become a philosophiæ doctor.

What I just got through saying will seem absurd to some. But there is an easy way to show that it is not at all absurd. After about three years of studies, many PhD students successfully hand in a dissertation. Now, give me an experimental grant (grant me an experiment). Take some of those students and give them two weeks of luxury on an island of their choice. Bring them home and give them a week to reread their dissertation. Now, give them 16 weeks to rewrite it as a defense of 255 claims to know something. Reward them with a return to that island for a month. That's 21 weeks (less than half a year) added to the program. It's luxurious, yes, but not absurdly so.

The lesson of this thought experiment is that, once you know what you want to say, you won't have any trouble writing your thesis. At bottom, you only have to say about 250 things. And you only have to claim to know those things.


Jonathan said...

I prefer to think of these claims in hierarchical fashion.

There will be one central claim in your dissertation: the thesis of your thesis. Each chapter will have a central claim of its own, so 4 subtheses, say.

Within each chapter, there will be three or four main points, so we're up to about 1 + 4 + 16.

Now each of these 16 subpoints will require about 15 paragraphs of support. Now we are talking about your 250-300 individual claims. Visualize them as a pyramid.

Thomas said...

I think we agree there. The point is that there's going to be about 220-250 paragraphs of "supporting" prose in your dissertation, and then another 25-50 paragraphs that guide those basic blocks of fact into an approximately pyramidical or icebergacious shape.

The supporting prose is interesting because it's presenting knowledge at the most detailed, most explicated level. It does not matter whether its in your theory section or in your analysis section. There's a level of detail that your claims, so to speak, "bottom out in".

I'm thinking about this in relation to your ideas about the capillary level. You said this should happen as a rare "show of strength". Probably true in a paper, and perhaps in general discourse (a lecture or teaching). But in a dissertation there's got to be capillaries, fibres, "meat", in bulk. Don't there?