Earlier this week, I posted at Mode for Caleb on Emma Dunham Kelley-Hawkins, the late nineteenth-century novelist long presumed to be an African American but now believed to have been white. The post followed up on Scott McLemee's column at Inside Higher Ed and Ralph Luker's post here at Cliopatria.

One of the things I suggested was this:"There seems to have been an enormous rush to judgment that the evidence provided by Holly Jackson, the Brandeis graduate student aforementioned, is straightforward proof about Kelley-Hawkins' race." That was a sloppy sentence, because it implied that I think Jackson might be wrong about Kelly-Hawkins' whiteness. McLemee, who generously linked to my post from his main page, responded thus:

As for CM's idea that there is a"rush to judgment," that certainly crossed my mind. But Holly Jackson actually makes a strong case for EDKH as white, while nobody has any solid evidence that she was black -- and when you read the scholarship, you notice people bending themselves out of shape to find some reason to think that she was. (I could have said more about that in the piece, including one case that verged on outright dishonesty, but that seemed like overdoing it.)

I agree completely that Jackson's evidence on the matter demolishes the traditional story about Kelly-Hawkins, since the traditional story is now apparently supported by no evidence at all. But if you look closer at my original statement on the"rush to judgment," what I wanted to question was not Jackson's particular evidence about Kelley-Hawkins, but rather the general presumption that certain kinds of historical evidence about race are inherently more decisive than others.

The idea that Kelley-Hawkins was black seems to have rested largely on the ambiguity of a photograph that appeared in one of her novels. Jackson's case for her whiteness rests, on the other hand, on family memories and documentary sources like the census. What I wanted to offer, however, was a cautionary warning that documentary sources about race are not necessarily less ambiguous than pictures. For instance, I wondered in the post about how nineteenth-century enumerators recorded" color" or"race." If, in some cases, determining a person's"race" was left to the enumerator's discretion, then a"W" in the census is not necessarily less ambiguous than a picture. It could simply record what a contemporary saw when he looked at Kelley-Hawkins. Does the census always offer us more information about race than meets the eye, or simply the same"information" provided from a contemporary perspective and packaged in documentary form? Such, at least, is a question I thought worth asking about Jackson's evidence. (An article by Martha Hodes in the February 2003 American Historical Review raised these questions about another Massachussetts family in the nineteenth century.)

At Coffee Grounds, Evan Roberts replied with an extremely helpful post about census procedures, which includes links to instructions that were given to enumerators. This information supplemented comments that Julie Meloni made on my original post, drawing on her own experiences using the census for genealogical research.

I highly recommend Evan's post, and I entirely agree with this sentence:"What I think is significant is that Emma Dunham Kelley-Hawkins and her ancestors are always described as white. That is firmer evidence of being white, whatever 'being' and 'white' mean." What I wanted to say in my post was that we should continue to add that final clause to whatever we say about race in the nineteenth century. This does not mean that we cannot make judgments, on the basis of contemporary descriptions, about the"race" of historical actors. But it means we must always keep in mind how ambiguous and arbitrary such descriptions about race could be and still are. And we must careful not to assume uncritically that certain kinds of evidence -- like the census -- somehow offer glosses of what"being" and"white" meant that are more determinate than other kinds of evidence -- like photographs. (Incidentally, Coffee Grounds also has two recent posts on the distinction between"quantitative" and"qualitative" sources that might be relevant here. I'm suggesting that even"quantitative" sources like the census were"qualitative.")

That's why I also suggested that the scholarship that has been done under the assumption that Kelley-Hawkins was not white should not simply be thrown out with the bathwater as a bunch of political misdirection. By presuming that Kelly-Hawkins was a black author creating aggresively white characters, that scholarship had to wrestle with what"being" and"white" meant. Without having read any of the scholarship that McLemee profiles, I think its insight that race can be malleable and strategically deployed is a valuable one. Although in this case scholars turn out to have been barking up the wrong tree, the bending themselves out of shape that they had to do is the kind of exercise that now makes us more limber as historians of race. In that limited sense, the underlying assumptions and theoretical underpinnings of that scholarship were not worthless. And it would be a shame if the conclusion that people drew from Kelley-Hawkins' case was that any scholarship on the cultural construction of"whiteness" and"blackness" is simply the product of postmodern mumbling and political correctness.