The art of writing history is making arguments about the dead. You’ll be dead one day, too, so please play fair, and remember: never condescend…. Every argument worth making begins with a question. Good questions come in all shapes and sizes. Very roughly, you can sort yours into two piles. One kind is more empirical (what happened?): “Why, on the eve of the American Revolution, did the painter John Singleton Copley decide to leave Boston?” The other is historiographical (what’s at stake in the debate among historians about what happened?)” “Have historians overstated the role of urban artisans in securing the repeal of the STamp Act?” The best, most rigorous, and most interesting scholarship answers both sorts of questions; it’s also much more fun to write, and to read.

Your question hasn’t been tattooed on your forehead. You can change it. Very likely, no one will even notice. If things are going well, you might decide, once you get into the research, that your question is bad, or even terrible. It might be the wrong question, but you can’t answer it. That’s fine; that’s excellent, actually; that’s what’s supposed to happen, so long as you think of a new question, or a chain of questions. A question isn’t a fish, a very wise historian once said; it’s a fishing license. It says what kind of fish you’re looking for, and where you’re going to put the boat. Never go fishing without a license. Once you’ve got that license, though, sail into the wide water, and cast your line. These instructions concern writing, not research. ‘How to Catch a Fish’ is a whole other handout…

Historians tend to write in both expository and narrative modes. In the wrting of history, a story without an argument fades into antiquarianism; and argument without a story risk pedantry. Rarely does any historian choose one mode to the exclusion of another, but how to balance these modes is a crucial choice. To answer that question about Copely, of course, you need to tell a good deal of the story of his life. You might begin, “John Singleton Copely was born in 1737 in a shake perched atop Boston’s longest wharf.” Or, you might want to begin with the claim: “Copely, long understood as a Tory, was loyal to nothing so much as his art.” Because you’re making an argument about the past, about something that happened, it almost inevitably has a natural narrative shape, a beginning, a middle, and an end. But an argument has a beginning, a middle, and an end, too. People lived and died. It’s nice to put them in your paper. But ideas are vital as well. So are institutions, theories, interpretations, and, above all, evidence. Most everything is vital, so long as the way you write about it is sufficiently animated, informed, and judicious.

If you don’t make an outline, you might as well throw your fish back into the water, guts and all. Where does your argument or your story begin, and where does it need to go, where has it got to end? What sequence of evidence best supports your claims? How and where will you engage both with what other scholars have written about your subject, and with broader interpretations of this period, or with theories about the past, or historical forces? … Show your reader your evidence, whenever you can — quote from primary documents — but don’t assume that those quotations explain themselves; offer a close reading. Your reader needs to see not only your evidence but also how you interpret it. Never discard evidence that counters your argument; show that evidence, too, and explain why your argument still stands. Do this with care, and respect. You can be convincing without being a bully. Bullies usually have lousy evidence, and very little to say.

When are you done? When you’ve stated your case, and finished your story. Lots of essays begin but never end; they merely stop. Make sure to end. It’s not over till there’s nothing left but the bones, and the smell of the ocean.