As a student of Native American history, I am well aware of the ways that the quantification of people, especially by the state, has been used to treat humans like objects, to strip them of humanity. This history is at the root of my discomfort with and skepticism of how digital/computational methods are sometimes used. It was interesting, then, to read Anelise H. Shrout’s talk from Digital Humanities 2017 because she argues that rather than reinforcing the “epistemic violence” of historical data in the archives, “DH methods, combined with insights from scholars who study marginalized people, can be used to undermine the inhumanity of that data.” Shrout gives three ways we can do something “with and against” the violence of quantification:
The first: as DH practitioners, we have to contend with the kinds of work that historical data creation enacted upon marginalized people and with the work that historical data producers thought they were doing. For my data, this means that I need to think about, for example, discrepancies between what we know about immigrants and what is revealed in the data, and what kind of explanatory power those discrepancies have.
The second: unfortunately, we have no archival records that describe the internal mechanisms of Bellevue. These processes were largely invisible to us, and remain largely archiavally invisible. But in looking at how thousands of immigrants moved through this system, we can start to see the institutional forces pushing on immigrants.
Finally, and this is what I want to close with, we can use quantitative methods to identify particular moments of contingency. Put another way, we can identify variables (each of which signifies one stage in immigrants’ passage through the almshouse) which significantly predict or are correlated with some other stage or experience, and then drill down into those moments, and imagine the ways in which immigrants within this system might have exercised agency.
I want to believe that digital methods can be used to challenge the violent power of quantification, but I don’t think I’ve reached a point where I can confidently assert that it’s possible. Nevertheless, I find the list above to be an incredibly useful starting point for myself in thinking about possible approaches to historical data. Anyone considering applying digital/computational methods to historical data on marginalized people would benefit from carefully considering these things.
Read the original post here.