I remember the first time I had to write a chapter of my dissertation. It was a daunting task. I had lots of sources, some ideas, and maybe even an argument, too. But I had no plan—no idea how to translate the thoughts in my head and the photos on my hard drive into an actual, coherent set of claims. I did what I knew how to do. I wrote what I thought was a passable outline, and then started writing.

That proved to be a mistake.

Weeks later and many thousands of words in, I was more confused than ever. I wasn’t sure what I was arguing, and even less convinced that I had organized that argument into anything close to a coherent narrative. I needed something better—more efficient, and that might force me to make an actual argument that wouldn’t confuse my reader.

Over time, I developed a new system for writing a chapter that emphasizes the upfront organizational work I outline here. The goal is to ensure that you engage in big-picture analytical or organizational thinking and verify that your evidence and claims align before you’ve spent a month and 8,000 words on a chapter, not after.

This chapter-writing system is distinguished by two features. First, it breaks down the process into distinct stages. Second, it forces you to think through much of the argumentation and structure in advance, shifting from big-picture thinking to detailed planning before you write a single sentence. Forcing yourself to clarify your argument and structure before the writing process—not after—makes the actual writing more streamlined and efficient. While it can be slow and frustrating at first, the time you invest in the beginning will result in a more organized and well-argued chapter.

Heavily edited 1828 letter found in the records of the U.S. Mint, proving that translating thoughts into written drafts has plagued writers since at least the nineteenth century. (Samuel Moore, “Draft of letter dated 7 February 1828,” Entry 1, General Correspondence, Record Group 104, Records of the US Mint 1792-1873, NARA, Philadelphia.) 

A few caveats: This is not the only or best way to write a chapter. Sometimes, you just need to write through problems to figure things out. But it is worthwhile to have a method, any method, as a place to start. This approach works best when you already have a general sense of what your chapter is going to argue (figuring out what to argue in the first place is a topic for a different blog post). The method outlined here is better for chapters where you have a general sense of the direction that you want to go, but are struggling to figure out how to get from a vague idea about argument and intervention to an actual chapter.

So, let’s begin.

Step 1: Read through all your primary-source notes related to the proposed chapter topic and collate all those sources into a single notebook or file. I use Evernote, so for me this entails creating a new notebook and importing sources over. Your specific way of doing this will probably depend on your system for taking notes. The point is to have all your research in one place, which will come in handy when you’re outlining.

The second effect of reading through all your primary-source notes is to remind you of what evidence you actually have. While this method doesn’t have a specific approach for helping you come up with an argument, reading through your sources is a good way to start generating ideas for an argument—and, in theory, will make sure that the argument reflects the evidence you’ve already found.

Once you know what your chapter is going to argue and how it fits in the literature (an equally thorny problem that I am hand-waving away here), it’s time to start thinking about the chapter itself.

Step 2: Write a single paragraph that distills the chapter’s argument. This is essentially an abstract, and it should answer the following questions: What is the chapter’s main thesis? What interventions are you making in the historiography?

This might seem simple, but it’s essential. Every part of the chapter is eventually going to come back to the statement. If you can’t articulate the chapter’s purpose in 200 words, then writing 10,000 words is going to result in a sprawling monster. This is the time to get clear on the main point of the chapter before you start drafting. Now, this paragraph is going to evolve as you work, and that’s okay. It’s a starting point, a single distillation of the chapter’s argument and intervention. And it forces you to pick only one argument.

Step 3: Write out a road map for the chapter. This is a somewhat longer statement, generally 500 or 800 words, that offers a narrative overview of the chapter’s structure. Think of this as the narrative arc of your chapter. What are its major sections or movements? What story does it tell? Why did the sections appear in a particular order? How do they build on each other?

It should take you quite a bit of time to write this statement—often a few days, including time to think things through. This statement is your first attempt to outline the chapter. You’re telling yourself the story of the chapter before you write it, and having to write it in a shorter statement will make you think through what sections you want to include and how they relate to each other. Think of it as an exercise, and not a piece of writing that others will read.

The goal is to force you to go through the mental process of thinking about the chapter in a systematic way. If the organization doesn’t work here, it will break down even more once you actually start writing the chapter. To that end, this statement should be fairly specific—what background do you need to offer up front? What will part 1 of the chapter look like? What order will it introduce sources? Or perhaps the chapter is a story, in which case, what are the major acts? Where will the chapter begin and end, and how will it get there?

Step 4: Now that you’ve got a narrative outline, it’s time to build a more traditional outline. In this case, you’re going to create a very specific kind of outline: one that lists the topic sentence for every paragraph in your chapter.

The math here is simple. For a 10,000-word chapter with 250 to 300 words per paragraph, you will write 33 to 40 paragraphs. And since each paragraph should have a topic sentence and one clear idea, your chapter will only contain 33 to 40 topic sentences and smaller points.

Still, this is where things get tricky. Confining yourself to the same number of paragraphs that you have in your final chapter  forces you to be economical with your argument. You’re going to quickly see where your argument is too sprawling or too thin. You’ll see where you’re trying to make points that aren’t important, or where you might need to break apart a more complicated set of arguments into multiple paragraphs. It may be that you’re tempted to spend eight paragraphs on something minor, and only two on something that’s really crucial to the larger argument. By forcing yourself to list out the main point of each paragraph as a topic sentence, you will  have to narrow things down.

Outlining in this way will also force you to make structural decisions. Does this claim need to appear before that one? Do you need to explain things in a certain order? Or maybe you’re jumping around too much. By writing at this level of granularity without actually writing the paragraphs themselves, you’ll start to get the flow right. I find that I generally spend a lot of time at this stage moving topic sentences around until they flow in a logical order. I have also found that I almost always want to include more than I realistically can to hit my word count.

Don’t forget: you get one point per paragraph. If a topic sentence is very long or contains two distinct ideas, you probably need two paragraphs. If you can’t articulate the point in a single sentence, you don’t know what that paragraph is doing.

Step 5: At this point, you should have an outline with 40 or so paragraphs, including your introduction, conclusion, and several sections. Now it’s time to go back to your primary and secondary sources. Find all the sources that you need for each paragraph and put them under the topic sentence. Be sure to include citations so that you can find your original sources. You may not use every source that you include here. Some will get cut during drafting. But the goal is to have everything you might need to write that paragraph already embedded in your outline. No more hunting through your files for the perfect quote you vaguely remember when you’re in the middle of writing.

The second purpose of this step is to make sure that you have the evidence you need to support your sub-arguments. If you can’t find evidence for a paragraph, maybe the point doesn’t belong, or you need to go back and adjust your claims. Luckily, you’re realizing this before you’ve spent weeks trying to write it out. If you find that you have a great deal of evidence for one paragraph and very little for the next few, you might want to think about rebalancing your argument.

Step 6: Finally, it’s time to write! Here’s the beauty of this approach. You already know exactly what you’re going to say when it’s time to say it. You have a topic sentence, you have your evidence, and you know how they all relate to the main point. In this way, you could drop in and write one paragraph at a time, and everything you need to write that paragraph is already there. Graduate student life is busy, and this will allow you to open your document, drop into your outline, find the next paragraph, and write it. Half an hour? You could knock out a paragraph without having to hunt things down. You’re not trying to hold the entire chapter’s argument in your head while writing, because you’ve already structured it out. You’re just executing—paragraph by paragraph.

That’s not to say that you can outline a chapter in this way, write it out, and never touch the outline again. You will absolutely revise during every part of this process. You’ll probably find new and better arguments, which will force you to restructure things. You may realize that your evidence doesn’t fit your claims or that there are new and better ways to explain things. You will reorganize all of this, and that’s good—a sign of intellectual evolution. But because you’ve done your hard organizational work up front, your revisions will hopefully be easier to make and require less wholesale restructuring than may have happened otherwise.

This method is much easier if you move beyond Microsoft Word. I personally prefer to use Scrivener and Evernote. Both are built for modular writing and make it easy to move things around. In Scrivener, for example, each paragraph can be its own document within the larger chapter structure, and multiple views allow you to see how it all fits together. There are other tools that would probably work well too, including Notion, OneNote, or maybe even a very well-organized Google document. The point is to have a system that lets you see both the big picture and the sentence level, and easily toggle between the two.

This approach works for a few reasons. First, it aligns the kind of work you’re doing with the decisions you need to be making at each stage. When you’re thinking about the overall argument, you don’t need to be worrying about sentence-level prose. When you’re getting into more granular levels of your argument at the outlining stage, the process of putting your sources underneath each topic sentence forces you to make sure that your evidence actually supports your claims. After you’ve figured out all these bigger questions about organization, structure, and evidence, you can turn your attention to that perfect paragraph prose. Writing isn’t one single process so much as a series of discrete tasks, and organizing the writing process accordingly makes it a more enjoyable activity and, hopefully, results in a better final product.