Yesterday, John Stauffer — Professor of English and African-American Studies at Harvard — published a piece for The Root entitled, “Yes, There Were Black Confederates. Here’s Why.” In it he argues that “between 3,000 and 6,000 served as Confederate soldiers. Another 100,000 or so blacks, mostly slaves, supported the Confederacy as laborers, servants and teamsters.”

It did not take very long for a number of Civil War historians to protest these conclusions, and to persuasively dismantle Stauffer’s argument piece by piece. First, Brooks Simpson got into it, and then Kevin Levin. Glenn David Brasher, whose book The Peninsula Campaign and The Necessity of Emancipation, details African-American involvement in that Union offensive, added his voice to the debate in the comments on Levin’s first post about the essay.

Now, this particular topic — black Confederates — has always provoked heated arguments, usually between historians and Lost Cause enthusiasts. But this latest fracas also brings up interesting questions about the uses (and misuses) of evidence in Civil War history, and about what actually constitutes “evidence” — and who gets to analyze it.

It is important to remember that Stauffer is a cultural studies scholar; he teaches and writes about nineteenth-century America through close-readings of written (and some visual) texts that produce cultural rhetorics and political action. I would imagine that he hit upon the topic of black Confederates through Frederick Douglass, whose writings are the subject of many of Stauffer’s articles and books.

In his essay for The Root Stauffer uses literature published in the North (in this case, Douglass’s speeches, newspaper articles, and also fugitive slave/freed peoples’ autobiographies and other wartime accounts) as his evidence. Now, if he had used these works to prove a point regarding ideas about southern black soldiers in circulation during the Civil War era, or about the production of the black Confederate image in Civil War literary and visual culture, the essay would have been more persuasive. But he did not. The problem is that he uses literature to try to prove a statistical (or, if you wish, a “historical” point): that there were thousands of southern slaves and freed people who “supported” the Confederacy....