Have a message. State it. Write nothing else.

Those are the first instructions to Demographic Research authors on their submission pages. I've heard variations on this theme before, but nothing as clearly stated as this. You might say that the Demographic Research editors had a message, stated it clearly, and wrote nothing else.

I started using this advice to reframe several papers that I have hanging around. What I have found in my first week using this advice is that the organization of papers has become easier. My writing becomes crisper. My papers are shorter. And I find them to be more compelling.

I noticed improvement in a couple of concrete areas. I realized that I have been summarizing the literature rather than using the literature to make an argument. This was frustrating to me because using the literature to make an argument is exactly what I tell students to do. I focused more on what I find rather than what I did. Again, something I advise students to do. Finally, my conclusions and implications were much narrower than they were previously. Although I hadn't been making wild speculations, not having a concrete message allowed me to wander. Readers, I presume (at least from the reviews I received) lost track of the central message at a critical point in the paper. This is something that I frequently mention on student papers.

As I was working on these papers, I realized that I have already written in this way. Almost every public health article I have written followed this model. Public health articles are shorter (3,500 words versus 8-10,000) and I think that the constraint forces more discipline on my writing (the downside is that public health articles can also easily leave out theoretical justifications and important methodological details).

I need to think along the same lines when writing sociology. Have a specific point and make it. The onus of making a theoretical contribution is not to be on par with Bourdieu, but to justify my approach. At least that's what I'm going with now. We'll see if the reviewers agree.